Communication method and apparatus

Abstract

The present application provides a communication method and an apparatus. The method comprises: a first apparatus obtains a target task; the first apparatus obtains a first capability from a shared capability library on the basis of the target task, wherein the shared capability library is used for storing capabilities, the capabilities in the shared capability library comprise the first capability, and the shared capability library supports sharing of the capabilities in the shared capability library by a plurality of first apparatuses; and the first apparatus uses the first capability to execute the target task. When processing similar target tasks, different first apparatuses can obtain, from a shared capability library, a first capability corresponding to the target tasks and use the first capability to execute the target tasks to obtain execution results, instead of separately processing the target tasks and separately generating capabilities for executing the target tasks. The above process can reduce unnecessary redundant operations, thereby reducing the time spent in task processing processes and improving the task processing efficiency of the first apparatuses.

Description

Communication method and apparatus

[0001] Cross-reference to Related Applications

[0002] This application claims priority to the Chinese Patent Application No. 202410906029.X, filed on July 5, 2024, and entitled "A Communication Method and Apparatus", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0003] The present application relates to the technical field of communication, and in particular, to a communication method and apparatus. BACKGROUND

[0004] With the rapid development of communication technology and artificial intelligence (AI) technology, in order to improve the intelligence of a terminal device, an artificial intelligence agent (AI Agent) can be configured for the terminal device. The terminal device can send the task to the AI Agent, and the AI Agent processes the task, which helps to improve the accuracy of the task processing result. At present, even if the obtained tasks are similar, different AI Agents still process their own tasks, which results in a large number of homogenization operations and low efficiency of task processing. SUMMARY

[0005] The present application provides a communication method and apparatus to improve the task processing efficiency of the AI Agent.

[0006] In a first aspect, an embodiment of the present application provides a communication method, and an execution subject of the method is a first device or a module or a chip in the first device. Here, the first device is taken as an example for description. The method comprises: the first device obtains a target task; the first device obtains a first capability from a shared capability library according to the target task; wherein the shared capability library is used to store capabilities, the capabilities in the shared capability library include the first capability, and the shared capability library supports multiple first devices sharing the capabilities in the shared capability library; and the first device executes the target task using the first capability.

[0007] In the method, different first devices can obtain a first capability corresponding to a target task from a shared capability library when processing the similar target task, and execute the target task using the first capability to obtain an execution result, instead of processing the target task and generating a capability for executing the target task respectively. The above process can reduce unnecessary homogenization operations, which is conducive to reducing the time spent in the task processing process and improving the task processing efficiency of the first device.

[0008] In a possible design, the first device obtaining the first capability from the shared capability library according to the target task can include: the first device determining at least one subtask corresponding to the target task; and the first device obtaining, from the shared capability library, at least one second capability corresponding to the at least one subtask respectively, where the first capability includes the at least one second capability corresponding to the at least one subtask respectively. In this design, the first device can split the target task into the at least one subtask, and obtain the first capability by obtaining the second capability corresponding to the subtask from the shared capability library, so that the above process can achieve the purpose of obtaining the first capability.

[0009] In a possible design, the first device determining the at least one subtask corresponding to the target task can include: the first device sending a first message to the second device, where the first message includes the target task, and the first message is used to request the at least one subtask corresponding to the target task; and the first device receiving the at least one subtask from the second device. In this design, the first device determines the at least one subtask corresponding to the target task through the second device, which can improve the efficiency of obtaining the at least one subtask, and meanwhile, since the second device can provide a large language model function, the at least one subtask determined through the second device is more accurate and reasonable.

[0010] In a possible design, the first message further includes information of a third capability, and the information of the third capability is used to determine the at least one subtask; and the method can further include: the first device sending a second message to a capability library, where the second message includes the target task, and the second message is used to request information of a capability corresponding to the target task; the capability library includes a local capability library corresponding to the first device and / or a shared capability library; the local capability library corresponding to the first device is used to store a capability, and the local capability library corresponding to the first device supports the first device corresponding to the first device to obtain the capability in the local capability library, but does not support other first devices to obtain the capability in the local capability library; and the first device receiving the information of the third capability from the capability library.

[0011] In a possible design, the information of the third capability includes at least one of the following:

[0012] an identifier of the third capability; or

[0013] description information of the third capability, where the description information of the third capability is used to describe a function of the third capability; or

[0014] a sharing manner of the third capability, where the sharing manner of the third capability is used to indicate a presentation manner of the third capability; or

[0015] parameter information of the third capability, where the parameter information of the third capability is used to indicate an input parameter required by using the third capability and / or to indicate an output parameter of the third capability; or

[0016] environment dependent information of the third capability, the environment dependent information of the third capability being used to indicate environment information relied on by the third capability; or

[0017] execution time of the third capability, the execution time of the third capability being used to indicate time required for executing the third capability; or

[0018] execution cost of the third capability, the execution cost of the third capability being used to indicate cost required for using the third capability; or

[0019] calling interface of the third capability, the calling interface of the third capability being used to implement the function of the third capability; or

[0020] source code of the third capability, the source code of the third capability being used to implement the function of the third capability.

[0021] source code permission policy of the third capability, the source code permission policy of the third capability being used to indicate whether the source code of the third capability allows modification / inheritance / derivation and / or whether other capabilities generated based on the source code of the third capability need to be added to the shared capability library.

[0022] In a possible design, the second message further includes a first constraint condition corresponding to the target task; and the third capability satisfies the first constraint condition.

[0023] In a possible design, the first constraint condition satisfies at least one of the following:

[0024] The first constraint condition is used to constrain execution time of the target task; or

[0025] The first constraint condition is used to constrain execution cost of the target task; or

[0026] The first constraint condition is used to constrain execution result of the target task; or

[0027] The first constraint condition is used to constrain an execution environment of the target task.

[0028] In a possible design, the at least one subtask includes a first subtask; and the first device obtaining the second capability corresponding to each of the at least one subtask from the shared capability library can include: the first device sending, to the second device, a third message, where the third message includes the first subtask, and the third message is used to request an identifier of the second capability corresponding to the first subtask; the first device receiving the identifier of the second capability corresponding to the first subtask from the second device; and the first device obtaining the second capability corresponding to the first subtask from the shared capability library according to the identifier of the second capability corresponding to the first subtask. Through the design, the first device can obtain the second capability used to execute the first subtask from the shared capability library.

[0029] In a possible design, the third message can further include information of a fourth capability corresponding to the first subtask, and the information of the fourth capability is used to determine the identifier of the second capability corresponding to the first subtask; the method can further include: sending, by the first device, a fourth message to the capability library, where the fourth message includes the first subtask, and the fourth message is used to request information of a capability corresponding to the first subtask; and receiving, by the first device, the information of the fourth capability from the capability library.

[0030] In a possible design, the fourth message can further include a second constraint condition corresponding to the first subtask, and the second constraint condition is determined according to the first constraint condition corresponding to the target task; and the fourth capability satisfies the second constraint condition.

[0031] In a possible design, the second constraint condition satisfies at least one of the following:

[0032] The second constraint condition is used to constrain an execution time of the first subtask, and the execution time of the first subtask is determined according to an execution time of the target task; or

[0033] The second constraint condition is used to constrain an execution cost of the first subtask, and the execution cost of the first subtask is determined according to an execution cost of the target task; or

[0034] The second constraint condition is used to constrain an execution result of the first subtask, and the execution result of the first subtask is determined according to an execution result of the target task; or

[0035] The second constraint condition is used to constrain an execution environment of the first subtask.

[0036] In a possible design, the first device can obtain the second capability corresponding to the first subtask from the shared capability library according to the identifier of the second capability corresponding to the first subtask, which can include: sending, by the first device, a fifth message to the shared capability library, where the fifth message includes the identifier of the second capability corresponding to the first subtask, and the fifth message is used to request the second capability corresponding to the first subtask; and receiving, by the first device, the second capability corresponding to the first subtask from the shared capability library. In this design, the first device can obtain the second capability used to execute the first subtask from the shared capability library.

[0037] In a possible design, the first device can obtain the second capability corresponding to the first subtask from the shared capability library according to the identifier of the second capability corresponding to the first subtask, which can include: obtaining, by the first device, the second capability corresponding to the first subtask from the information of the fourth capability according to the identifier of the second capability corresponding to the first subtask, where the information of the fourth capability is from the shared capability library. In this design, the first device can obtain the second capability used to execute the first subtask from the information of the fourth capability, without accessing the shared capability library again, which is beneficial to improving the efficiency of obtaining the second capability.

[0038] In a possible design, the first device can obtain the second capability corresponding to the first subtask according to the identifier of the second capability corresponding to the first subtask, which can include: the first device can obtain the second capability corresponding to the first subtask according to the identifier of the second capability corresponding to the first subtask and authorization information, where the authorization information is used to indicate authorization of the first device to obtain the second capability corresponding to the first subtask.

[0039] In a possible design, the method can further include: the first device can send a sixth message to the third device, where the sixth message includes the identifier of the second capability corresponding to the first subtask, and the sixth message is used to request authorization information of the second capability corresponding to the first subtask; and the first device can receive the authorization information of the second capability corresponding to the first subtask from the third device. In this design, the first device can quickly obtain the authorization information of the second capability corresponding to the first subtask, which is beneficial to improving the efficiency of obtaining the authorization information of the second capability corresponding to the first subtask.

[0040] In a possible design, the method can further include: the first device can send a seventh message to the shared capability library, where the seventh message includes the second capability corresponding to any of the at least one subtask, and the seventh message is used to indicate that the second capability corresponding to any of the at least one subtask is added to the shared capability library. In this design, the first device can add the generated capability to the shared capability library, which is convenient for other first devices to obtain and use the capability from the shared capability library, and avoids unnecessary homogenization operations.

[0041] In a possible design, the method can further include: the first device can determine a fifth capability according to the second capabilities corresponding to the plurality of subtasks of the at least one subtask, where the fifth capability is used to implement the target task or is used to implement part of the target task; and the first device can send a seventh message to the shared capability library, where the seventh message includes the fifth capability, and the seventh message is used to indicate that the fifth capability is added to the shared capability library. In this design, the first device can add the generated capability to the shared capability library, which is convenient for other first devices to obtain and use the capability from the shared capability library, and avoids unnecessary homogenization operations.

[0042] In a possible design, the first apparatus determines the fifth capability according to the second capabilities corresponding to the multiple subtasks in the at least one subtask, which can include: the first apparatus sending an eighth message to the second apparatus, where the eighth message includes the second capabilities corresponding to the multiple subtasks in the at least one subtask, and the eighth message is used to indicate that the fifth capability is generated according to the capabilities corresponding to the multiple subtasks in the at least one subtask; and the first apparatus receiving the fifth capability from the second apparatus. In this design, the second apparatus generates the fifth capability instead of the first apparatus, which can reduce power consumption of the first apparatus. In addition, the fifth capability generated by the second apparatus is more complete and reasonable.

[0043] In a possible design, the seventh message further includes at least one of the following:

[0044] an identifier of the first apparatus; or

[0045] description information of the fifth capability, where the description information of the fifth capability is used to describe a function of the fifth capability; or

[0046] a sharing manner of the fifth capability, where the sharing manner of the fifth capability is used to indicate a presentation manner of the fifth capability; or

[0047] parameter information of the fifth capability, where the parameter information of the fifth capability is used to indicate an input parameter required for using the fifth capability and / or to indicate an output parameter of the fifth capability; or

[0048] environment dependency information of the fifth capability, where the environment dependency information of the fifth capability is used to indicate environment information on which the fifth capability depends; or

[0049] an execution time of the fifth capability, where the execution time of the fifth capability is used to indicate a time required for executing the fifth capability; or

[0050] an execution cost of the fifth capability, where the execution cost of the fifth capability is used to indicate a cost required for using the fifth capability; or

[0051] a calling interface of the fifth capability, where the calling interface of the fifth capability is used to implement the function of the fifth capability; or

[0052] source code of the fifth capability, where the source code of the fifth capability is used to implement the function of the fifth capability.

[0053] a source code permission policy of the fifth capability, where the source code permission policy of the fifth capability is used to indicate whether the source code of the fifth capability allows modification / inheritance / derivation and / or whether other capabilities generated based on the source code of the fifth capability need to be added to the shared capability library.

[0054] In a possible design, the first device sends the seventh message to the shared capability library, which can include: in a case where the verification of the fifth capability is passed, the first device sends the seventh message to the shared capability library.

[0055] In a possible design, the verification of the fifth capability by the first device includes at least one of the following:

[0056] verification of an execution time of the fifth capability; or

[0057] verification of a dependent environment of the fifth capability; or

[0058] verification of code corresponding to the fifth capability.

[0059] In a second aspect, an embodiment of the present application provides a communication method, an execution subject of the method is a shared capability library or a module or chip in the shared capability library, and the shared capability library is taken as an execution subject for example. The method includes: receiving, by the shared capability library, a task from a first device; and sending, by the shared capability library, a first capability corresponding to the task to the first device, the first capability being used to execute the task, the shared capability library being used to store capabilities, the capabilities in the shared capability library including the first capability, and the shared capability library supporting multiple first devices sharing the capabilities in the shared capability library.

[0060] In the method, the shared capability library can provide the capabilities corresponding to the task for the multiple first devices, so that different first devices do not need to process the task and generate the capabilities used to execute the target task respectively when processing similar tasks, which helps to reduce unnecessary homogenization operations, reduce time spent in the task processing process, and improve the task processing efficiency of the first device.

[0061] In a possible design, the shared capability library receives the task from the first device, which can include: the shared capability library receives a target task from the first device; and the shared capability library sends the first capability corresponding to the task to the first device, which can include: the shared capability library sends the first capability corresponding to the target task to the first device. In this design, after obtaining the target task, the first device does not need to split the target task, but directly obtains the first capability corresponding to the target task from the shared capability library, which can avoid the splitting process of the target task and the process of generating executable code for each subtask, reduce unnecessary homogenization operations, and be beneficial to reducing time spent in the above processing process and improving the efficiency of obtaining an execution result of the target task.

[0062] In a possible design, the shared capability library receives the task from the first device, which can include that the shared capability library receives at least one subtask corresponding to the target task from the first device; and the shared capability library sends the first capability corresponding to the task to the first device, which can include that the shared capability library sends second capabilities corresponding to the at least one subtask respectively to the first device, and the first capability includes the second capabilities corresponding to the at least one subtask respectively. In this design, the first device can split the target task into the at least one subtask, and obtain the second capabilities corresponding to the subtasks from the shared capability library to achieve the purpose of obtaining the first capability, so that the above process can achieve the purpose of obtaining the first capability.

[0063] In a possible design, the at least one subtask includes a first subtask; the shared capability library sends the second capabilities corresponding to the at least one subtask respectively to the first device, which can include that the shared capability library receives a fifth message from the first device, where the fifth message includes an identifier of the second capability corresponding to the first subtask, and the fifth message is used to request the second capability corresponding to the first subtask; and the shared capability library sends the second capability corresponding to the first subtask to the first device. In this design, the first device can obtain the second capability used to perform the first subtask from the shared capability library.

[0064] In a possible design, the method can further include that the shared capability library deducts an execution cost of the second capability corresponding to the first subtask from an account corresponding to the first device.

[0065] In a possible design, the shared capability library receives a seventh message from the first device, where the seventh message includes the second capability corresponding to any of the at least one subtask, and the seventh message is used to indicate that the second capability corresponding to any of the at least one subtask is added to the shared capability library; and the shared capability library adds the second capability corresponding to any of the at least one subtask to the shared capability library. In this design, the first device can add the generated capability to the shared capability library, so as to facilitate other first devices to obtain and use the capability from the shared capability library, and avoid unnecessary homogenization operations.

[0066] In a possible design, the method can further include that the shared capability library issues a reward corresponding to the second capability corresponding to any of the at least one subtask to an account corresponding to the first device.

[0067] In a possible design, the method further includes: receiving, by the shared capability library, a seventh message from the first device, where the seventh message includes the fifth capability, the fifth capability is generated according to the second capability corresponding to each of the at least one subtask, and the seventh message is used to indicate that the fifth capability is added to the shared capability library; and adding, by the shared capability library, the fifth capability to the shared capability library. In this design, the first device can add the generated capability to the shared capability library, so as to facilitate other first devices to obtain and use the capability from the shared capability library, and avoid unnecessary homogenization operations.

[0068] In a possible design, the method further includes: issuing, by the shared capability library, a reward corresponding to the fifth capability to an account corresponding to the first device.

[0069] In a possible design, the method further includes: performing, by the shared capability library, capability optimization on the capabilities in the shared capability library.

[0070] In a possible design, the capability optimization on the capabilities in the shared capability library includes at least one of the following: capability deletion, capability merging, capability replacement, and capability reconstruction. In this design, the quality of the capabilities in the shared capability library can be continuously improved by performing capability optimization on the capabilities in the shared capability library, and the execution efficiency of the task can be improved.

[0071] In a third aspect, the present application provides a communication apparatus, which can implement any of the methods in the first aspect and the second aspect. The communication apparatus can be implemented by hardware, or by hardware executing corresponding software. The hardware or software includes one or more units or modules corresponding to the above functions.

[0072] In a possible design, the communication apparatus includes a processor configured to support the communication apparatus to perform the corresponding functions performed by the first device or the shared capability library in the above methods. The communication apparatus can further include a memory coupled to the processor, which stores program instructions and data necessary for the communication apparatus. Optionally, the communication apparatus further includes an interface circuit, which is configured to support the communication apparatus to communicate with a terminal device or other devices.

[0073] In a possible design, the communication apparatus includes corresponding functional modules, which are respectively configured to implement the steps in the above methods. The functions can be implemented by hardware, or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

[0074] In a possible design, a communication apparatus includes a processing unit and a communication unit, which can perform the corresponding functions in the method examples described above, details of which can be referred to the description of the method in any of the first aspect to the second aspect, which will not be repeated here.

[0075] In a fourth aspect, an embodiment of the present application further provides a communication apparatus, which includes a module / unit for performing any of the methods provided in the first aspect to the second aspect.

[0076] In a fifth aspect, an embodiment of the present application provides a communication apparatus, including a memory and a processor; the processor is configured to execute computer programs or instructions stored in the memory, so that the communication apparatus implements any of the methods provided in the first aspect to the second aspect.

[0077] In a sixth aspect, an embodiment of the present application further provides a computer readable storage medium, which includes a computer program, and when the computer program is run on a communication apparatus, the computer readable storage medium implements any of the methods provided in the first aspect to the second aspect.

[0078] In a seventh aspect, an embodiment of the present application further provides a computer program product, which, when run on a communication apparatus, makes the communication apparatus implement any of the methods provided in the first aspect to the second aspect.

[0079] In an eighth aspect, a chip is provided, which includes a processor and can further include a memory, and is configured to implement any of the methods provided in the first aspect to the second aspect. The chip can be composed of a chip, or can include a chip and other discrete devices.

[0080] In a ninth aspect, a communication system is provided, including: a first apparatus and a shared capability library; the first apparatus is configured to implement the method in the first aspect and any possible implementation manner in the first aspect; and the shared capability library is configured to implement the method in the second aspect and any possible implementation manner in the second aspect.

[0081] In a tenth aspect, a computer program product is provided, which, when read and executed by a computer, makes the computer implement any of the methods provided in the first aspect to the second aspect.

[0082] These aspects or other aspects of the present application will become apparent in the following description of the embodiments with reference to the images. BRIEF DESCRIPTION OF DRAWINGS

[0083] FIG. 1 is a structural schematic diagram of a communication system according to an embodiment of the present application.

[0084] FIG. 2 is a structural diagram of a communication system according to an embodiment of the present application;

[0085] FIG. 3 is a flow diagram of a communication method according to an embodiment of the present application;

[0086] FIG. 4 is a flow diagram of a method for obtaining a first capability according to an embodiment of the present application;

[0087] FIG. 5 is a flow diagram of a method for obtaining a first capability according to an embodiment of the present application;

[0088] FIG. 6 is a flow diagram of a method for obtaining at least one subtask according to an embodiment of the present application;

[0089] FIG. 7 is a flow diagram of a method for obtaining information of a third capability according to an embodiment of the present application;

[0090] FIG. 8 is a flow diagram of a method for obtaining a second capability corresponding to a first subtask according to an embodiment of the present application;

[0091] FIG. 9 is a flow diagram of a method for obtaining information of a fourth capability according to an embodiment of the present application;

[0092] FIG. 10 is a flow diagram of a method for obtaining authorization information of a second capability corresponding to a first subtask according to an embodiment of the present application;

[0093] FIG. 11 is a flow diagram of a method for adding a capability according to an embodiment of the present application;

[0094] FIG. 12 is a flow diagram of a method for determining a fifth capability according to an embodiment of the present application;

[0095] FIG. 13 is a structural diagram of a communication apparatus according to an embodiment of the present application;

[0096] FIG. 14 is a structural diagram of a communication apparatus according to an embodiment of the present application. DETAILED DESCRIPTION

[0097] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. The terms "first", "second", and corresponding terms of reference numbers in the specification and claims of the present application and the drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged under appropriate circumstances, and this is only a distinguishing way used in the description of the embodiments of the present application to describe the objects with the same attribute. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, so that the processes, methods, systems, products or equipment containing a series of units do not have to be limited to those units, but can include other units not clearly listed or inherent to these processes, methods, products or equipment.

[0098] In the following, some terms in the embodiments of the present application are explained and described, so as to facilitate the understanding of the skilled in the art.

[0099] A large language model (LLM) is a deep learning model trained using a large amount of text data, which can generate natural language text or understand the meaning of language text. Large language models can handle a variety of natural language tasks such as text classification, question answering, and dialogue, and are an important way to artificial intelligence.

[0100] Prompt information, also known as guidance information, is a piece of prompt text used to guide large language models to generate specific types of output in large language models. By designing and optimizing a set of prompt information, large language models can be made to perform specific tasks.

[0101] An artificial intelligence agent, also known as an intelligent agent or an agent, is an intelligent entity that can perceive the environment, make decisions, and perform actions, designed to efficiently execute and process complex tasks through natural language interaction with large language models.

[0102] The technical solutions disclosed in the present application can be applied to a 5th generation mobile communication technology (5G) communication system, for example, a 5G new radio (NR) system, and can also be applied to various application scenarios of the 5G communication system, such as enhanced mobile broadband (eMBB), ultra reliable low latency communication (URLLC), and enhanced machine-type communication (eMTC). The technical solutions disclosed in the present application can also be applied to various communication systems in future evolution, such as future communication systems, space-aerospace-sea-land integrated communication systems. The technical solutions disclosed in the present application can also be applied to communication between base stations and base stations, communication between terminal devices and terminal devices, vehicle networking, Internet of Things, industrial Internet, satellite communication, and the like, for example, can be applied to device-to-device (D2D), vehicle-to-everything (V2X), and machine-to-machine (M2M) communication systems.

[0103] FIG. 1 is a schematic diagram of a communication system to which embodiments of the present application are applicable. As shown in FIG. 1, the communication system 10 includes one or more network devices 20 and one or more terminal devices 30. The interface between the network device and the terminal device can be a Uu interface (or air interface), and the network device 20 and the terminal device 30 can perform data transmission through the air interface resource. The terminal device can access a data network (DN) through the network device. The components of the communication system are described below.

[0104] (1) Terminal device

[0105] The terminal device includes a device providing voice and / or data connectivity to a user, e.g., a device having wireless transceiver functionality that can be disposed in a terrestrial environment, including indoor or outdoor, handheld, wearable or vehicle mounted; can be disposed on a water surface (e.g., a ship or the like); can also be disposed in air (e.g., on an airplane, a balloon or a satellite or the like). The terminal device can be a tag, a mobile phone, a tablet computer, a computer with wireless transceiver functionality, a virtual reality (VR) terminal, an augmented reality (AR) terminal, a wireless terminal in industrial control, a vehicle-mounted terminal, a wireless terminal in self driving, a wireless terminal in remote medical, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, a wireless terminal in smart home, a wearable terminal, or the like. The embodiments of the present application do not limit the application scenarios. The terminal device can also be referred to as a terminal, a user equipment (UE), an access terminal, a vehicle-mounted terminal, an industrial control terminal, a UE unit, a UE station, a mobile station, a mobile station, a remote station, a remote terminal, a mobile device, a UE terminal, a wireless communication device, a UE agent or a UE apparatus, etc. The terminal device can be fixed or mobile. It can be understood that all or part of the functions of the terminal device in the present application can also be implemented by software functions running on hardware, or by virtualized functions instantiated on a platform (e.g., a cloud platform). The terminal device in the present application can be a terminal for 5G or a terminal for future communication networks, which is not limited in the present application.

[0106] (2) Network device

[0107] The network device can include an access network device and a core network device.

[0108] An access network device is a device that provides a terminal device with a wireless communication function. The access network device includes, but is not limited to, a future base station (g nodeB, gNB) in 5G, an evolved node B (eNB), a radio network controller (RNC), a node B (NB), a base station controller (BSC), a base transceiver station (BTS), a home base station (for example, a home evolved nodeB, or a home node B, HNB), a baseband unit (BBU), a transmitting and receiving point (TRP), a transmitting point (TP), a mobile switching center, and the like.

[0109] The access network device and the terminal device can be fixed in position or movable. The access network device and the terminal device can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; can also be deployed on the water surface; and can also be deployed on an airplane, a balloon, and a man-made satellite in the air. Embodiments of the present application do not limit the application scenarios of the access network device and the terminal device.

[0110] The core network device includes part or all of the following network elements: a unified data management (UDM) network element, a unified data repository (UDR) network element, a network exposure function (NEF) network element, an application function (AF) network element, a policy control function (PCF) network element, an access and mobility management function (AMF) network element, a tag management function (TMF) network element, a session management function (SMF) network element, a user plane function (UPF) network element, and a network repository function (NRF) network element.

[0111] The access and mobility management function network element is a control plane network element provided by an operator network, responsible for access control and mobility management of terminal devices accessing the operator network, for example, including functions such as mobile state management, allocation of user temporary identity, authentication and authorization of users, etc. In 5G, the access and mobility management function network element can be an AMF network element, and in future communication networks, the access and mobility management function network element can still be an AMF network element, or have other names, which are not limited in the present application.

[0112] The user plane function network element is a gateway provided by an operator, which is a gateway for communication between the operator network and the DN. The user plane function network element includes functions related to the user plane, such as data packet routing and transmission, packet detection, service usage reporting, quality of service (QoS) processing, lawful monitoring, uplink packet detection, and downlink data packet storage. In 5G, the user plane network element can be a UPF network element, and in future communication networks, the user plane network element can still be a UPF network element, or have other names, which are not limited in the present application.

[0113] In the network architecture shown in FIG. 1, in order to improve the intelligence of the terminal device, an intelligent agent can be set for the terminal device, wherein the intelligent agent can be located in the terminal device or in the network device. The intelligent agent located in the network device can provide related services to the terminal device, which is not limited herein.

[0114] Taking the network architecture shown in FIG. 1 as an example, the network architecture includes two terminal devices, and the intelligent agent corresponding to each terminal device is located in the network device. The network architecture shown in FIG. 1 includes terminal device 1 and terminal device 2. User 1 can use terminal device 1, and the intelligent agent set for terminal device 1 is intelligent agent 1. User 2 can use terminal device 2, and the intelligent agent set for terminal device 2 is intelligent agent 2.

[0115] User 1 inputs a target task in terminal device 1, that is, "help me predict the future weather in Nanjing", and terminal device 1 sends the target task to intelligent agent 1. Intelligent agent 1 splits the target task and obtains three subtasks, which are as follows:

[0116] 1. Obtain a weather webpage list;

[0117] 2. Read the Nanjing weather webpage from the weather webpage list;

[0118] 3. Read the weather information of Nanjing in the next 15 days from the Nanjing weather webpage.

[0119] The intelligent agent 1 obtains the three sub-tasks, generates executable codes corresponding to the three sub-tasks respectively, executes the three sub-tasks in sequence using the executable codes corresponding to the three sub-tasks respectively, and thus obtains the weather information of Nanjing in the next 15 days.

[0120] Similarly, the user 2 inputs a target task, i.e., "help me predict the weather of Shenzhen in the future", into the terminal device 2, the terminal device 2 sends the target task to the intelligent agent 2, the intelligent agent 2 splits the target task and obtains three sub-tasks as follows:

[0121] 1. Obtain a weather webpage list;

[0122] 2. Read a Shenzhen weather webpage from the weather webpage list;

[0123] 3. Read weather information of Shenzhen in the next 15 days from the Shenzhen weather webpage.

[0124] The intelligent agent 2 obtains the three sub-tasks, generates executable codes corresponding to the three sub-tasks respectively, executes the three sub-tasks in sequence using the executable codes corresponding to the three sub-tasks respectively, and thus obtains the weather information of Shenzhen in the next 15 days.

[0125] As can be seen from the above, the intelligent agent 1 or the intelligent agent 2 needs to split the target task to obtain multiple sub-tasks and generate executable codes corresponding to each sub-task when processing the target task, and the above processing process takes a relatively long time.

[0126] In addition, the target tasks to be processed by the intelligent agent 1 and the intelligent agent 2 are similar, and the processes of generating executable codes by the intelligent agent 1 and the intelligent agent 2 are also similar, which causes unnecessary homogenization operation and also leads to a relatively long time spent in the above processing process.

[0127] Therefore, the embodiments of the present application provide a communication method to solve the problem that the intelligent agent takes a relatively long time to process the target task.

[0128] The communication method provided by the embodiments of the present application is applicable to the communication system shown in FIG. 2, and the communication system at least includes a first device and a shared capability library (SCL), wherein the first device can communicate with the shared capability library, and the first device can also communicate with a terminal device. Optionally, the communication system shown in FIG. 2 can also include a local capability library (LCL) corresponding to the first device, a second device and a third device. The communication connection relationship between the first device, the local capability library corresponding to the first device, the shared capability library, the second device and the third device in the communication system can refer to FIG. 2, and the communication connection manner between the first device, the local capability library corresponding to the first device, the shared capability library, the second device and the third device in the communication system can be wired connection or wireless connection, which is not limited here.

[0129] The first device, the local capability library corresponding to the first device, the shared capability library, the second device and the third device in the communication system shown in FIG. 2 are explained in detail as follows:

[0130] 1) The first device is a device providing intelligent agent function. The first device can be a virtual device or a physical device, which is not limited here. The first device can obtain a target task from a terminal device and return the result of the target task to the terminal device.

[0131] One first device can correspond to one terminal device, that is, one first device can receive a target task from one terminal device. In this case, the first device can be located in the terminal device, can be located in a network device, or can be an independent network element structure, which is not limited here.

[0132] One first device can correspond to multiple terminal devices, that is, one first device can receive target tasks from multiple terminal devices. In this case, the first device can be located in a network device, or the first device can be an independent network structure, which is not limited here.

[0133] 2) The local capability library corresponding to the first device is used to store capabilities, one local capability library corresponds to one first device, and one local capability library supports the first device corresponding to the local capability library to obtain the capabilities in the local capability library, and does not support other first devices to obtain the capabilities in the local capability library.

[0134] For each first device, there can be a local capability library corresponding to each first device, or there can be no local capability library corresponding to each first device. In the case that there is a local capability library corresponding to each first device, each first device can correspond to one local capability library, or multiple local capability libraries, which is not limited here.

[0135] The local capability library corresponding to the first device can be located in a terminal device, can be located in a network device, or can be a separate structure, without limitation.

[0136] The capability stored in the local capability library can be understood as source code for implementing a task, or can be understood as an executable file for implementing a task, without limitation.

[0137] It should be understood that the capability in the local capability library corresponding to the first device can be generated and added by the first device, can be generated by the local capability library through capability optimization on the capability in the local capability library, or can be a capability possessed by any network device in the communication network, such as a computing capability or a sensing capability, without limitation.

[0138] 3) The shared capability library is used to store capabilities, and the shared capability library supports multiple first devices sharing the capabilities in the shared capability library.

[0139] The shared capability library can be located in a terminal device, can be located in a network device, or can be a separate structure, without limitation.

[0140] The capability stored in the shared capability library can be understood as source code for implementing a task, or can be understood as an executable file for implementing a task, without limitation.

[0141] It should be understood that the capability in the shared capability library can be generated and added by any first device, can be generated by the shared capability library through capability optimization on the capability in the shared capability library, can be added by a capability provider (CP), or can be a capability possessed by any network device in the communication network, without limitation.

[0142] 4) The second device is a device providing a large language model function. The second device can be a virtual device or a physical device, without limitation. The first device can send a target task to the second device, the second device can split the target task through a large language model to obtain at least one subtask, and the first device can send the at least one subtask to the first device.

[0143] One second device can correspond to one first device, i.e. one second device can receive target tasks from one first device. In this case, if the first device is located in a terminal device, the second device can be located in the terminal device, or in a network device, or be a standalone network element structure. If the first device is located in a network device, the second device can be located in the network device, or be a standalone network element structure. If the first device is a standalone network element structure, the second device can be located in a network device, or be a standalone network element structure.

[0144] One second device can correspond to multiple first devices, i.e. one second device can receive target tasks from multiple first devices. In this case, if the first device is located in a network device, the second device can be located in the network device, or be a standalone network element structure. If the first device is a standalone network element structure, the second device can be located in a network device, or be a standalone network element structure.

[0145] 5) The third device is a device that can provide authorized information of capabilities. The third device can be a virtual device or a physical device, which is not limited herein.

[0146] The third device can receive an authorization request message sent by the first device, wherein the authorization request message is used to request authorized information of capabilities of any task, and the third device can send the authorized information of capabilities of the task to the first device after receiving the authorization request message.

[0147] One third device can correspond to one first device, i.e. one third device can receive an authorization request message from one first device. In this case, if the first device is located in a terminal device, the third device can be located in the terminal device, or in a network device, or be a standalone network element structure. If the first device is located in a network device, the third device can be located in the network device, or be a standalone network element structure. If the first device is a standalone network element structure, the third device can be located in a network device, or be a standalone network element structure.

[0148] One third device can correspond to multiple first devices, i.e. one third device can receive authorization request messages from multiple first devices. In this case, if the first device is located in a network device, the third device can be located in the network device, or be a standalone network element structure. If the first device is a standalone network element structure, the third device can be located in a network device, or be a standalone network element structure.

[0149] It should be understood that the first device, the local capability library corresponding to the first device, the shared capability library, the second device and the third device can be integrated in the same device, or can be located in different devices respectively, or can be partially located in the same device, which is not limited herein.

[0150] With reference to the system architecture diagram shown in FIG. 2, the method embodiment shown in FIG. 3 is exemplarily given to make a detailed description of the technical solution of the present application. It should be noted that FIG. 3 is a schematic flowchart of the method embodiment of the present application, which shows the detailed communication steps or operations of the method, but these steps or operations are only examples, and other operations or variations of the various operations in FIG. 3 can also be performed by the embodiments of the present application. In addition, the various steps in FIG. 3 can be performed in different order from that shown in FIG. 3, and it is possible that not all operations in FIG. 3 are performed. The communication method shown in FIG. 3 is performed by the first device and the shared capability library in FIG. 2.

[0151] S301, the first device acquires a target task.

[0152] For example, a user can input a target task in a terminal device, where the description manner of the target task can be text, voice, image, etc., which is not limited herein. For the convenience of the following description, the description manner of the target task is taken as text in the following description.

[0153] In the case where the first device is located in the terminal device, the first device can directly acquire the target task from the terminal device; in the case where the first device is located in a network device or is an independent network element structure, the terminal device can send the target task to the first device.

[0154] Optionally, the first device can also acquire the task constraint condition of the target task. Wherein, the description manner of the task constraint condition can be text, voice, image, etc., which is not limited herein. For the convenience of the following description, the description manner of the task constraint condition of the target task is taken as text in the following description.

[0155] Exemplarily, the first device can acquire the task constraint condition of the target task by any one of the following implementation manners A1, A2, or A3.

[0156] Implementation manner A1, the first device generates the task constraint condition according to the target task.

[0157] In the embodiments of the present application, after acquiring the target task, the first device can automatically generate the task constraint condition of the target task.

[0158] Implementation manner A2, the first device acquires the task constraint condition of the target task from a task constraint configuration.

[0159] In a possible implementation, the first device stores a task constraint configuration, and the task constraint configuration can include task constraint conditions corresponding to a plurality of tasks, where the plurality of tasks in the task constraint configuration include the target task. After obtaining the target task, the first device can obtain the task constraint condition corresponding to the target task from the task constraint configuration.

[0160] In an implementation A3, the first device receives the task constraint condition of the target task from the terminal device, where the task constraint condition of the target task is input by a user in the terminal device.

[0161] In the embodiments of the present application, when the first device is located in the terminal device, the first device can directly obtain the task constraint condition of the target task from the terminal device; when the first device is located in the network device or is an independent network element structure, the terminal device can send the task constraint condition of the target task to the first device.

[0162] The task constraint condition of the target task is explained in detail as follows, and the task constraint condition of the target task satisfies at least one of the following conditions:

[0163] 1) The task constraint condition is used to constrain the execution time of the target task.

[0164] For example, the task constraint condition can include an upper limit value or a time range of the execution time of the target task. For example, the task constraint condition can include 30 seconds, indicating that the target task is completed within 30 seconds; or the task constraint condition can include 10 seconds to 30 seconds, indicating that the target task is completed within 10 seconds to 30 seconds.

[0165] 2) The task constraint condition is used to constrain the execution cost of the target task.

[0166] For example, the task constraint condition can include an upper limit value or a cost range of the execution cost of the target task. The execution cost can be in the form of points, currency, etc., which is not limited herein. For example, the task constraint condition can include 100 points, indicating that the target task is completed within 100 points; or the task constraint condition can include 100 points to 200 points, indicating that the target task is completed within 100 points to 200 points.

[0167] 3) The task constraint condition is used to constrain the execution result of the target task.

[0168] For example, the task constraint condition can include form information of the execution result of the target task. For example, the task constraint condition can include that the execution result of the target task is displayed in a chart form, or the execution result of the target task is displayed in a text form, or the execution result of the target task is rounded, etc.

[0169] In addition, the task constraint condition of the target task can also constrain other aspects of the target task, which is not limited herein.

[0170] It should be understood that the user can input the target task and the task constraint condition of the target task simultaneously in the terminal device, and the first device obtains the target task and the constraint condition of the target task from the terminal device simultaneously. The user can also input the target task in the terminal device first, and then the first device obtains the target task from the terminal device. The user inputs the task constraint condition of the target task in the terminal device again, and then the first device obtains the task constraint condition of the target task from the terminal device.

[0171] In addition, the embodiments of the present application do not limit the input form of the target task and the task constraint condition. For example, the user can input a description text in the terminal device, and the first device obtains the description text and analyzes the description text to obtain the target task and the task constraint condition of the target task after obtaining the description text.

[0172] For example, the user inputs a description text in the terminal device, the content of the description text is "help me predict the future weather in Nanjing, need to be completed within 30 seconds, the consumption of this time cannot exceed 100 points, and the prediction result is displayed in the form of a chart and text", and the first device obtains the description text and analyzes the description text to obtain the target task "help me predict the future weather in Nanjing" and the task constraint condition "need to be completed within 30 seconds, the consumption of this time cannot exceed 100 points, and the prediction result is displayed in the form of a chart and text" corresponding to the target task after obtaining the description text.

[0173] S302, the first device obtains a first capability from the shared capability library according to the target task. The first capability can be used to execute the target task and obtain an execution result of the target task.

[0174] In the embodiments of the present application, the first device can obtain the first capability through any of the following implementation manners B1 or B2.

[0175] Implementation manner B1 includes the following steps shown in FIG. 4:

[0176] S401, the first device sends a ninth message to the shared capability library; correspondingly, the shared capability library receives the ninth message from the first device.

[0177] In the embodiments of the present application, the ninth message includes the target task, the ninth message is used to request the first capability corresponding to the target task, or the ninth message is used to indicate to obtain the first capability corresponding to the target task.

[0178] Optionally, the ninth message can further include a first constraint condition, wherein the first constraint condition is determined according to a task constraint condition of the target task and / or environment information of the first device.

[0179] In a possible implementation, the environment information of the first device includes software environment information and / or hardware environment information of the first device. The software environment information of the first device can include an operating system version number corresponding to the first device, and the hardware configuration information of the first device can include a central processing unit (CPU) model and a network card model corresponding to the first device.

[0180] The first constraint condition is explained in detail as follows. The first constraint condition can satisfy at least one of the following:

[0181] 1) The first constraint condition is used to constrain an execution time of the target task.

[0182] In the embodiments of the present application, the detailed description of the first constraint condition used to constrain the execution time of the target task can refer to the detailed description of the task constraint condition used to constrain the execution time of the target task in the foregoing, which is not repeated here.

[0183] 2) The first constraint condition is used to constrain an execution cost of the target task.

[0184] In the embodiments of the present application, the detailed description of the first constraint condition used to constrain the execution cost of the target task can refer to the detailed description of the task constraint condition used to constrain the execution cost of the target task in the foregoing, which is not repeated here.

[0185] 3) The first constraint condition is used to constrain an execution result of the target task.

[0186] In the embodiments of the present application, the detailed description of the first constraint condition used to constrain the execution result of the target task can refer to the detailed description of the task constraint condition used to constrain the execution result of the target task in the foregoing, which is not repeated here.

[0187] 4) The first constraint condition is used to constrain an execution environment of the target task.

[0188] In the embodiments of the present application, the first constraint condition can include environment information of the first device, and the environment information of the first device is used to constrain the execution environment of the target task.

[0189] In addition, the first constraint condition can further include other aspects of constraints, which are not limited here.

[0190] S402, the shared capability library determines a first capability corresponding to the target task.

[0191] In the embodiments of the present application, the shared capability library can store at least one capability and information of each capability, can also store at least one capability and an encoding vector corresponding to information of each capability, and can also store at least one capability, information of each capability and an encoding vector of information of each capability, which are not limited herein. The encoding vector corresponding to the information of the capability can be generated by using an embedding model or generated by using other manners, which are not limited herein.

[0192] In a possible implementation, the information of the capability in the shared capability library comprises at least one of the following:

[0193] 1) an identifier of the capability.

[0194] In the embodiments of the present application, the identifier of the capability is used to distinguish different capabilities, and the identifier of the capability can be a name of the capability, a number of the capability, a storage path of the capability, etc., which are not limited herein. The identifier of the capability can be determined by the shared capability library for each capability, or can be determined by the first device or the CP providing the capability, which are not limited herein.

[0195] 2) description information of the capability.

[0196] In the embodiments of the present application, the description information of the capability is used to describe the function of the capability. The description information of the capability can be a text, which is not limited herein.

[0197] 3) a sharing mode of the capability.

[0198] In the embodiments of the present application, the sharing mode of the capability is used to indicate a presentation mode of the capability. The sharing mode of the capability can be a source code sharing mode or an interface sharing mode, wherein the interface sharing mode can comprise an API sharing mode and an Agent interface sharing mode.

[0199] If the sharing mode of the capability is the source code sharing mode, it means that the source code of the capability needs to be acquired, and then the capability can be used.

[0200] If the sharing mode of the capability is the interface sharing mode, it means that the source code of the capability does not need to be acquired, and the calling interface of the capability can be directly called, and then the capability can be used.

[0201] If the interface sharing mode is the API sharing mode, the calling interface of the capability is an application program interface.

[0202] If the interface sharing mode is the Agent interface sharing mode, the calling interface of the capability is an Agent interface.

[0203] 4) parameter information of the capability.

[0204] In the embodiments of the present application, the parameter information of the capability is used to indicate the input parameters required by the capability and / or to indicate the output parameters of the capability. In the case that the sharing mode of the capability is the source code sharing mode or the interface sharing mode, the parameter information of the capability can be set.

[0205] 5) Environment dependent information of the capability.

[0206] In the embodiments of the present application, the environment dependent information of the capability is used to indicate the environment information on which the capability depends. The environment dependent information of the capability includes software environment information on which the capability depends and / or hardware configuration information on which the capability depends, etc. The software environment information on which the capability depends can include an operating system version number on which the capability depends, etc. The hardware configuration information on which the capability depends can include a CPU model on which the capability depends, a network card model on which the capability depends, etc.

[0207] 6) Execution time of the capability.

[0208] In the embodiments of the present application, the execution time of the capability is used to indicate the time required for executing the capability.

[0209] 7) Execution cost of the capability.

[0210] In the embodiments of the present application, the execution cost of the capability is used to indicate the cost required for using the capability. The execution cost of the capability can be in the form of points, in the form of money, etc., which are not limited herein.

[0211] 8) Invocation interface of the capability.

[0212] In the embodiments of the present application, the first device can implement the function of the capability by invoking the invocation interface of the capability. Generally, the invocation interface of the capability can be an invocation interface corresponding to a callable code, which is not limited herein.

[0213] In the embodiments of the present application, in the case that the sharing mode of the capability is the interface sharing mode, the information of the capability includes the invocation interface of the capability.

[0214] In the case that the interface sharing mode is the API sharing mode, the invocation interface of the capability is an application program interface.

[0215] In the case that the interface sharing mode is the Agent interface sharing mode, the invocation interface of the capability is an Agent interface.

[0216] 9) Source code of the capability.

[0217] In the embodiments of the present application, the first device can implement the function of the capability by invoking the source code of the capability. Generally, the source code of the capability can be an executable code, which is not limited herein.

[0218] In the embodiments of the present application, in the case that the sharing mode of the capability is the source code sharing mode, the information of the capability comprises a source code of the capability.

[0219] 10) a source code permission policy of the capability.

[0220] In the embodiments of the present application, the source code permission policy of the capability is used to indicate whether the source code of the capability allows modification / inheritance / derivation and / or whether other capabilities generated based on the source code of the capability need to be added to the shared capability library. It should be understood that the source code permission policy of the capability allowing inheritance / derivation means that other source codes of capabilities can use the source code of the capability and add other functional codes on the basis of the source code of the capability.

[0221] In a possible implementation, in the case that the source code permission policy of the capability indicates that the source code of the capability does not allow modification / inheritance / derivation, the source code of the capability is only allowed to be used and cannot be modified / inherited / derived.

[0222] In a possible implementation, in the case that the source code permission policy of the capability indicates that other capabilities generated based on the source code of the capability need to be added to the shared capability library, the other capabilities generated based on the source code of the capability need to be added to the shared capability library. In the case that the source code permission policy of the capability indicates that other capabilities generated based on the source code of the capability do not need to be added to the shared capability library, the other capabilities generated based on the source code of the capability can be added to the shared capability library or can not be added to the shared capability library.

[0223] In the embodiments of the present application, the shared capability library can determine the first capability corresponding to the target task by any of the following implementation modes C1 or C2.

[0224] In implementation mode C1, the shared capability library first calculates the similarity between the information of each capability in the shared capability library and the target task in the ninth message, and regards the capability with a similarity greater than a first similarity threshold as the first capability. The first capability can comprise one or more capabilities in the shared capability library.

[0225] In the above implementation mode C1, the similarity between the information of each capability and the target task can be determined by the following method: the shared capability library first encodes the target task to obtain a first encoding vector; then, the shared capability library calculates the similarity between the encoding vector of the information of each capability and the first encoding vector as the similarity between the information of each capability and the target task.

[0226] The first capability can be quickly determined by the above implementation mode C1, and the determined first capability can implement the target task.

[0227] In an implementation C2, the shared capability library can calculate a similarity between information of each capability in the capability library and the target task and the first constraint condition in the ninth message, and take a capability with a similarity greater than a first similarity threshold as the first capability. The first capability can include one or more capabilities in the shared capability library.

[0228] In the implementation C2 described above, the similarity between information of each capability and the target task and the first constraint condition in the ninth message can be determined by the following method: the shared capability library first encodes the target task and the first constraint condition to obtain a second encoding vector; then, the shared capability library calculates a similarity between an encoding vector of information of each capability and the second encoding vector as the similarity between information of each capability and the target task and the first constraint condition in the ninth message.

[0229] The implementation C2 described above can accurately determine the first capability, and the determined first capability can not only achieve the target task but also satisfy the first constraint condition.

[0230] In S403, the shared capability library sends the first capability to the first device.

[0231] In the embodiments of the present application, the shared capability library can also connect accounts corresponding to a plurality of first devices, and can increase or decrease account balances in the accounts corresponding to the plurality of first devices. It should be understood that the account corresponding to the first device can be established, managed and maintained by a third party institution.

[0232] In a possible implementation, when the shared capability library sends the first capability to the first device, the shared capability library can obtain an execution cost of the first capability from information of the first capability, and deduct the execution cost of the first capability from the account corresponding to the first device. The shared capability library can also add the deducted execution cost of the first capability to the account corresponding to the first device that provides the first capability.

[0233] Through the implementation B1 described above, after obtaining the target task, the first device does not need to split the target task, but directly obtains the first capability corresponding to the target task from the shared capability library, which can avoid the splitting process of the target task and the process of generating capabilities for each subtask, thereby improving the efficiency of obtaining the first capability. In addition, different first devices can obtain the first capability corresponding to the target task from the shared capability library when processing similar target tasks, instead of processing the target task and generating capabilities for executing the target task respectively, which can reduce unnecessary homogenization operations, and is conducive to reducing the time spent in the task processing process and improving the task processing efficiency of the first device.

[0234] Before performing the above-mentioned embodiment B1, the first device can send a ninth message to the local capability library corresponding to the first device, and the ninth message is used to request the first capability corresponding to the target task. If the local capability library can determine the first capability corresponding to the target task from the local capability library, the local capability library can directly send the first capability to the first device. Then, the first device does not need to perform the above-mentioned embodiment B1. If the local capability library cannot determine the first capability corresponding to the target task from the local capability library, the first device performs the above-mentioned embodiment B1. Since the first device acquires the capability from the local capability library more efficiently, in the case that the first capability corresponding to the target task exists in the local capability library, the above-mentioned method can quickly acquire the first capability.

[0235] It should be understood that the method for the first device to acquire the first capability from the local capability library can refer to the method for the first device to acquire the first capability from the shared capability library, which will not be described herein.

[0236] Embodiment B2 includes the following steps shown in FIG. 5:

[0237] S501, the first device determines at least one subtask corresponding to the target task.

[0238] The embodiments of the present application provide two possible implementation manners for determining at least one subtask corresponding to the target task.

[0239] The first possible implementation manner is that the first device splits the target task to obtain at least one subtask.

[0240] The second possible implementation manner includes the following steps shown in FIG. 6:

[0241] S601, the first device sends a first message to the second device; correspondingly, the second device receives the first message from the first device.

[0242] In the embodiments of the present application, the first message includes the target task, and the first message is used to request at least one subtask corresponding to the target task, or the first message is used to indicate to acquire at least one subtask corresponding to the target task.

[0243] Optionally, the first message can further include a first constraint condition, and the detailed description of the first constraint condition will not be repeated here.

[0244] Optionally, the first message can further include information of a third capability, wherein the information of the third capability can be provided by the local capability library and / or the shared capability library corresponding to the first device.

[0245] In a possible implementation manner, the information of the third capability includes at least one of the following:

[0246] 1) an identifier of the third capability;

[0247] 2) description information of the third capability, the description information of the third capability being used for describing a function of the third capability;

[0248] 3) a sharing mode of the third capability, the sharing mode of the third capability being used for indicating a presentation mode of the third capability;

[0249] 4) parameter information of the third capability, the parameter information of the third capability being used for indicating an input parameter required by using the third capability and / or indicating an output parameter of the third capability;

[0250] 5) environment dependency information of the third capability, the environment dependency information of the third capability being used for indicating environment information on which the third capability depends;

[0251] 6) execution time of the third capability, the execution time of the third capability being used for indicating a time required for executing the third capability;

[0252] 7) execution cost of the third capability, the execution cost of the third capability being used for indicating a cost required for using the third capability;

[0253] 8) a calling interface of the third capability, the calling interface of the third capability being used for implementing the function of the third capability;

[0254] 9) source code of the third capability, the source code of the third capability being used for implementing the function of the third capability;

[0255] 10) a source code permission policy of the third capability, the source code permission policy of the third capability being used for indicating whether the source code of the third capability allows modification / inheritance / derivation and / or whether other capabilities generated based on the source code of the third capability need to be added to the shared capability library.

[0256] It should be understood that detailed descriptions about the information of the third capability described above can refer to the foregoing detailed descriptions about the information of the capability in the shared capability library, and will not be described herein.

[0257] The embodiment of the present application provides a method for a first device to acquire information of a third capability, including the following steps as shown in FIG. 7:

[0258] S701, the first device sends a second message to the capability library; correspondingly, the capability library receives the second message from the first device.

[0259] In the embodiment of the present application, the second message includes a target task, and the second message is used for requesting information of a capability corresponding to the target task, or the second message is used for indicating to acquire the information of the capability corresponding to the target task.

[0260] In the embodiments of the present application, the capability library includes a local capability library corresponding to the first device and / or a shared capability library. The local capability library corresponding to the first device can only connect to an account corresponding to the first device and can increase or decrease the account balance in the account corresponding to the first device. The shared capability library can connect to multiple accounts corresponding to multiple first devices and can increase or decrease the account balance in the multiple accounts corresponding to the multiple first devices.

[0261] S702, the capability library determines information of the third capability.

[0262] In the embodiments of the present application, the method for determining the third capability in S702 can refer to the method for determining the first capability in S402, which will not be repeated here.

[0263] S703, the capability library sends the information of the third capability to the first device; correspondingly, the first device receives the information of the third capability from the capability library.

[0264] It should be understood that, in the case where the information of the third capability sent by the capability library to the first device includes the calling interface of the third capability or the source code of the third capability, the capability library can deduct the execution cost of the third capability from the account corresponding to the first device according to the execution cost of the third capability.

[0265] Through the method shown in FIG. 7, the first device can quickly determine the information of the third capability, which facilitates subsequent subtask planning for the target task according to the information of the third capability.

[0266] In the above S601, the first device can generate the first message through any one of the following embodiments D1, D2, D3 or D4.

[0267] Embodiment D1, the first device can directly add the target task to the first message, or can first perform prompt word processing on the target task to obtain prompt information, and then add the prompt information to the first message. The prompt information generated by the first device each time the target task is processed can be completely the same, or can be partially the same, which will not be limited here.

[0268] For example, the target task is "help me predict the future weather in Nanjing", the first device can directly add the target task to the first message, so the first message can directly include the information "help me predict the future weather in Nanjing"; or the first device can first perform prompt word processing on the target task to obtain prompt information such as "you are a weather forecast intelligent assistant, your target task is to predict the weather in Nanjing in the next month, please give the subtask planning of the target task", and then the first device adds the prompt information to the first message, and the first message can include the information "you are a weather forecast intelligent assistant, your target task is to predict the weather in Nanjing in the next month, please give the subtask planning of the target task".

[0269] In an embodiment D2, the first device can directly add the target task and the first constraint condition to the first message, or first perform prompt word processing on the target task and the first constraint condition to obtain prompt information, and then add the prompt information to the first message.

[0270] For example, the target task is "help me predict the future weather in Nanjing", and the first constraint condition includes "need to be completed within 30 seconds", "consumption cannot exceed 100 points", and "prediction results are displayed in the form of charts and texts". The first device can directly add the target task and the first constraint condition to the first message, so the first message can include the following content: "target task: help me predict the future weather in Nanjing; first constraint condition: need to be completed within 30 seconds, consumption cannot exceed 100 points, and prediction results are displayed in the form of charts and texts".

[0271] Or, the first device can first perform prompt word processing on the target task and the first constraint condition to obtain prompt information such as "you are a weather forecast intelligent assistant, your target task is to predict the weather in Nanjing in the next month, your constraint condition is to complete it within 30 seconds, and the consumed points cannot exceed 100, and the prediction results need to be displayed in the form of charts and texts, please give the subtask planning of the target task", and then the first device adds the prompt information to the first message, and the first message can include the above prompt information.

[0272] In an embodiment D3, the first device can directly add the target task and the information of the third capability to the first message, or first perform prompt word processing on the target task and the information of the third capability to obtain prompt information, and then add the prompt information to the first message.

[0273] For example, the target task is "help me predict the future weather in Nanjing", and the third capability information includes three identifiers of the third capabilities that can be used, i.e., "web search capability", "web reading capability", and "file reading and writing capability". The first device can directly add the target task and the third capability information to the first message, so the first message can include the following content: "target task: help me predict the future weather in Nanjing; third capability information: web search capability, web reading capability, and file reading and writing capability".

[0274] Alternatively, the first device can first perform prompt word processing on the target task and the third capability information to obtain prompt information such as "you are a weather forecast intelligent assistant, your target task is to predict the future weather in Nanjing for one month, you can call the web search capability, web reading capability, and file reading and writing capability, please give the subtask planning of the target task". The first device then adds the prompt information to the first message, and the first message can include the above prompt information.

[0275] In an embodiment D4, the first device can directly add the target task, the first constraint condition, and the third capability information to the first message, or first perform prompt word processing on the target task, the first constraint condition, and the third capability information to obtain prompt information, and then add the prompt information to the first message.

[0276] For example, the target task is "help me predict the future weather in Nanjing", the first constraint condition includes "need to be completed within 30 seconds", "consumption cannot exceed 100 points", and "prediction results are displayed in the form of charts and texts", and the third capability information includes three identifiers of the third capabilities that can be used, i.e., "web search capability", "web reading capability", and "file reading and writing capability". The first device can directly add the target task, the first constraint condition, and the third capability information to the first message, so the first message can include the following content: "target task: help me predict the future weather in Nanjing; first constraint condition: need to be completed within 30 seconds, consumption cannot exceed 100 points, and prediction results are displayed in the form of charts and texts; third capability information: web search capability, web reading capability, and file reading and writing capability".

[0277] Alternatively, the first device can first perform prompt word processing on the information of the target task, the first constraint condition and the third capability, and obtain prompt information such as "You are a weather forecast intelligent assistant, your target task is to predict the weather in Nanjing in the next month, your constraint condition is to complete within 30 seconds, and the consumed points shall not exceed 100, and the prediction result needs to be displayed in the form of a chart and text. Please give the subtask planning of the target task, and you can call the web search capability, web reading capability and file reading and writing capability. Please give the subtask planning of the target task". The first device then adds the prompt information to the first message. The first message can include the above prompt information.

[0278] In S602, the second device determines at least one subtask corresponding to the target task.

[0279] In the embodiments of the present application, for the four different implementation manners of generating the first message, the second device can determine at least one subtask corresponding to the target task through any one of the following implementation manners E1, E2, E3 or E4.

[0280] In implementation manner E1, in the case where the first message includes the target task or the prompt information generated based on the target task, after obtaining the target task in the first message, the second device can input the target task into the large language model, and the large language model outputs at least one subtask.

[0281] For example, in the case where the first message includes "You are a weather forecast intelligent assistant, your task target is to predict the weather in Nanjing in the next month, please give the subtask planning of the target task", the second device can determine the following four subtasks:

[0282] 1) Obtain a list of weather web pages;

[0283] 2) Read the Nanjing weather web page from the list of weather web pages;

[0284] 3) Read the weather information of Nanjing in the next 15 days from the Nanjing weather web page;

[0285] 4) Based on the weather information of Nanjing in the next 15 days, predict the weather information of Nanjing in the next month.

[0286] It should be understood that the execution result obtained after the at least one subtask is executed in sequence is equivalent to the execution result obtained after the target task is executed.

[0287] In the above implementation manner E1, the large language model directly determines at least one subtask according to the target task, which can quickly and efficiently generate at least one subtask and improve the efficiency of generating subtasks.

[0288] In the first message includes the target task and the first constraint condition or the prompt information generated based on the target task and the first constraint condition, the second device can input the target task and the first constraint condition into the large language model after obtaining the target task and the first constraint condition in the first message, and output at least one subtask and a second constraint condition corresponding to each subtask from the large language model. The second constraint condition corresponding to each subtask is determined according to the first constraint condition corresponding to the target task.

[0289] In the embodiments of the present application, the at least one subtask includes a first subtask, which can be any subtask in the at least one subtask. Taking the second constraint condition corresponding to the first subtask as an example, the second constraint condition corresponding to the first subtask is explained in detail as follows. The second constraint condition corresponding to the first subtask can satisfy at least one of the following conditions:

[0290] 1) The second constraint condition is used to constrain the execution time of the first subtask.

[0291] In the embodiments of the present application, the execution time of the first subtask is determined according to the execution time of the target task.

[0292] The embodiments of the present application provide the following two possible implementation manners to determine the execution time of the first subtask.

[0293] In the first possible implementation manner, the execution time of the first subtask can be determined according to the execution time of the target task and the number of subtasks. That is, the execution time of the first subtask is equal to the ratio of the execution time of the target task to the number of subtasks.

[0294] For example, if the first constraint condition includes a time upper limit value of 30 seconds for the execution time of the target task, the second device splits the target task to obtain three subtasks, and the time upper limit value of the execution time of each subtask is 10 seconds.

[0295] In the second possible implementation manner, the execution time of the first subtask can be determined according to the execution time of the target task and the complexity of the first subtask. The complexity of the first subtask can be determined by the large language model in the second device.

[0296] For example, if the first constraint condition includes a time upper limit value of 30 seconds for the execution time of the target task, the second device splits the target task to obtain three subtasks, namely subtask 1, subtask 2 and subtask 3. The complexity of subtask 1 and subtask 2 is higher, and the complexity of subtask 3 is lower. Therefore, the time upper limit value of the execution time of subtask 1 is 12 seconds, the time upper limit value of the execution time of subtask 2 is 12 seconds, and the time upper limit value of the execution time of subtask 1 is 6 seconds.

[0297] 2) The second constraint condition is used to constrain the execution cost of the first subtask.

[0298] In the embodiments of the present application, the execution cost of the first subtask is determined according to the execution cost of the target task.

[0299] The embodiments of the present application provide the following two possible implementation manners to determine the execution cost of the first subtask.

[0300] In the first possible implementation manner, the execution cost of the first subtask can be determined according to the execution cost of the target task and the number of subtasks. That is, the execution cost of the first subtask is equal to the ratio of the execution cost of the target task to the number of subtasks.

[0301] In the second possible implementation manner, the execution cost of the first subtask can be determined according to the execution cost of the target task and the complexity of the first subtask. The complexity of the first subtask can be determined by the large language model in the second device.

[0302] 3) The second constraint condition is used to constrain the execution result of the first subtask.

[0303] In the embodiments of the present application, the execution result of the first subtask is determined according to the execution result of the target task.

[0304] In the case where the first subtask is the last subtask corresponding to the target task, the form information of the execution result of the first subtask is the same as the form information of the execution structure of the target task.

[0305] 4) The second constraint condition is used to constrain the execution environment of the first subtask.

[0306] In the embodiments of the present application, the execution environment of the target task can be used as the execution environment of the first subtask.

[0307] For example, in the case where the first message includes "You are a weather forecast intelligent assistant, your target task is to predict the weather in Nanjing in the next month, your constraint condition is to complete within 30 seconds, the consumed points should not exceed 100, and the prediction result needs to be displayed in the form of charts and text, please give the subtask planning of the target task", the second device can determine the following 7 subtasks:

[0308] 1) Obtain a list of weather web pages; the corresponding second constraint condition is to complete within 2 seconds and consume no more than 10 points;

[0309] 2) Read the Nanjing weather web page from the list of weather web pages; the corresponding second constraint condition is to complete within 2 seconds and consume no more than 15 points;

[0310] 3) read the weather information of Nanjing in the next 15 days from the Nanjing weather webpage; the corresponding second constraint condition is to complete within 5 seconds and consume no more than 20 points;

[0311] 4) predict the weather information of Nanjing in the next month based on the weather information of Nanjing in the next 15 days; the corresponding second constraint condition is to complete within 10 seconds and consume no more than 30 points;

[0312] 5) draw a curve chart of the weather information of Nanjing in the next month; the corresponding second constraint condition is to complete within 5 seconds and consume no more than 20 points;

[0313] 6) summarize the weather information of Nanjing in the next month in words; the corresponding second constraint condition is to complete within 5 seconds and consume no more than 20 points;

[0314] 7) write the curve chart and the word summary of the weather information of Nanjing in the next month into a file and save it; the corresponding second constraint condition is to complete within 1 second and consume no more than 5 points.

[0315] In the above embodiment E2, the large language model determines at least one subtask according to the target task and the first constraint condition, and the determined at least one subtask all has a corresponding second constraint condition, so as to facilitate subsequent determination of the corresponding ability of each subtask according to the second constraint condition corresponding to each subtask.

[0316] In the case where the first message includes the information of the target task and the third ability or the prompt information generated based on the information of the target task and the third ability, the second device can input the information of the target task and the third ability into the large language model after obtaining the information of the target task and the third ability in the first message, and output at least one subtask by the large language model. Wherein, the large language model refers to the information of the third ability to plan subtasks for the target task to obtain at least one subtask, and part or all of the subtasks in the at least one subtask can be executed by the third ability.

[0317] For example, in the case where the first message includes "You are a weather forecast intelligent assistant, your target task is to predict the weather of Nanjing in the next month, you can call the web search ability, web reading ability, file reading and writing ability, please give the subtask planning of the target task", the second device can determine the following 4 subtasks:

[0318] 1) obtain a weather webpage list;

[0319] 2) read the Nanjing weather webpage from the weather webpage list;

[0320] 3) read the weather information of Nanjing in the next 15 days from the weather webpage of Nanjing;

[0321] 4) predict the weather information of Nanjing in the next month based on the weather information of Nanjing in the next 15 days.

[0322] Among the 4 sub-tasks determined by the second device, the sub-task of "obtaining the weather webpage list" can be executed by the webpage search capability in the third capability, the sub-task of "reading the weather webpage of Nanjing from the weather webpage list" can be executed by the webpage reading capability in the third capability, and the sub-task of "reading the weather information of Nanjing in the next 15 days from the weather webpage of Nanjing" can be executed by the file reading and writing capability in the third capability.

[0323] In the above embodiment E3, the large language model refers to the information of the third capability to plan sub-tasks for the target task to obtain at least one sub-task, so that the subsequent sub-tasks can be directly executed by using the third capability, and it is possible to avoid obtaining the capability corresponding to the sub-task from the capability library or generating the capability for each sub-task again, which is beneficial to improve the efficiency of executing the sub-tasks.

[0324] In the above embodiment E3, the large language model refers to the information of the third capability to plan sub-tasks for the target task to obtain at least one sub-task, so that the subsequent sub-tasks can be directly executed by using the third capability, and it is possible to avoid obtaining the capability corresponding to the sub-task from the capability library or generating the capability for each sub-task again, which is beneficial to improve the efficiency of executing the sub-tasks.

[0325] For example, in the case where the first message includes "You are a weather forecast intelligent assistant, your target task is to predict the weather of Nanjing in the next month, your constraint condition is to complete within 30 seconds, and the consumed points should not exceed 100, the prediction result needs to be displayed in the form of charts and texts, and you can call the webpage search capability, webpage reading capability, and file reading and writing capability, please give the sub-task planning of the target task", the second device can determine the following 7 sub-tasks:

[0326] 1) Obtain the weather webpage list; the corresponding second constraint condition is to complete within 2 seconds and consume no more than 10 points;

[0327] 2) Read the weather webpage of Nanjing from the weather webpage list; the corresponding second constraint condition is to complete within 2 seconds and consume no more than 15 points;

[0328] 3) reading the weather information of Nanjing in the next 15 days from the Nanjing weather webpage; the corresponding second constraint condition is to complete within 5 seconds and consume no more than 20 points;

[0329] 4) predicting the weather information of Nanjing in the next month based on the weather information of Nanjing in the next 15 days; the corresponding second constraint condition is to complete within 10 seconds and consume no more than 30 points;

[0330] 5) drawing a curve chart of the weather information of Nanjing in the next month; the corresponding second constraint condition is to complete within 5 seconds and consume no more than 20 points;

[0331] 6) summarizing the weather information of Nanjing in the next month in words; the corresponding second constraint condition is to complete within 5 seconds and consume no more than 20 points;

[0332] 7) writing the curve chart and the word summary corresponding to the weather information of Nanjing in the next month into a file and saving; the corresponding second constraint condition is to complete within 1 second and consume no more than 5 points.

[0333] In the above embodiment E4, the large language model determines at least one subtask according to the target task, the first constraint condition and the information of the third capability, so as to facilitate subsequent direct use of the third capability to execute the subtask. The determined at least one subtask all has a corresponding second constraint condition, so as to facilitate subsequent determination of the capability corresponding to each subtask according to the second constraint condition corresponding to each subtask.

[0334] Optionally, after determining the at least one subtask corresponding to the target task, the second device can also access the capability library to obtain the second capability corresponding to each subtask.

[0335] It should be understood that the second capability corresponding to one or more subtasks in the at least one subtask comes from the shared capability library, and the following will not be repeated.

[0336] S603, the second device sends the at least one subtask to the first device; correspondingly, the first device receives the at least one subtask from the second device.

[0337] Optionally, in the case that the second device acquires the at least one subtask and the second capability corresponding to each of the at least one subtask, the second device can send the at least one subtask and the second capability corresponding to each of the at least one subtask to the first device, and accordingly, the first device receives the at least one subtask and the second capability corresponding to each of the at least one subtask from the second device, and then the first device does not need to perform S502 and the subsequent steps. In the above method, the second device can acquire the at least one subtask and the second capability corresponding to each of the at least one subtask and send them to the first device, which is beneficial to improve the efficiency of acquiring the second capability.

[0338] In the above method shown in FIG. 6, the second device determines the at least one subtask corresponding to the target task, which can improve the efficiency of acquiring the at least one subtask, and the at least one subtask determined by the large language model in the second device is more accurate and reasonable.

[0339] After the first device determines the at least one subtask corresponding to the target task in S501, the first device can determine whether there is a third capability that can be used to execute the subtask according to the information of the third capability. If there is, the third capability is directly used to execute the subtask; if not, the capability corresponding to the subtask is acquired from the local capability library and / or the shared capability library corresponding to the first device.

[0340] For the convenience of the following description, the following examples are taken from the first device acquiring the capability corresponding to the subtask from the shared capability library, and the first device acquiring the capability corresponding to the subtask from the local capability library can refer to the first device acquiring the capability corresponding to the subtask from the shared capability library, which will not be described hereinafter.

[0341] S502, the first device acquires the second capability corresponding to each of the at least one subtask from the shared capability library.

[0342] In the embodiment of the present application, the first capability includes the second capability corresponding to each of the at least one subtask. It can be understood that the execution result obtained by using the second capability corresponding to each of the at least one subtask to execute the subtask is equivalent to the execution result obtained by using the first capability to execute the target task.

[0343] In the above S502, the first device can acquire the second capability corresponding to each of the at least one subtask through the steps shown in FIG. 8. The following takes the first subtask in the at least one subtask as an example to explain in detail how to acquire the second capability corresponding to the first subtask, and the first subtask is any subtask in the at least one subtask.

[0344] S801, the first device sends a third message to the second device; accordingly, the second device receives the third message from the first device.

[0345] In the embodiments of the present application, the third message comprises the first subtask, and the third message is used to request an identifier of the second capability corresponding to the first subtask, or the third message is used to indicate to obtain the identifier of the second capability corresponding to the first subtask.

[0346] Optionally, the third message can further comprise a second constraint condition corresponding to the first subtask. Details of the second constraint condition are not described herein.

[0347] Optionally, the third message can further comprise information used to indicate one or more of the following: a subtask executed before the first subtask, a result of the subtask executed before the first subtask, or a target task. The information can be used to determine the identifier of the second capability corresponding to the first subtask.

[0348] Optionally, the third message can further comprise fourth capability information corresponding to the first subtask, wherein the fourth capability information is used to determine the identifier of the second capability corresponding to the first subtask, and the fourth capability information can be provided by a local capability library and / or a shared capability library corresponding to the first device. Details of the fourth capability information can be referred to the description of the third capability information, and are not described herein.

[0349] For the convenience of the following description, the fourth capability information is taken as an example from the shared capability library in the following description. The fourth capability information from the local capability library corresponding to the first device can be referred to the fourth capability information from the shared capability library, and is not described herein.

[0350] The embodiments of the present application provide a method for a first device to obtain fourth capability information, comprising the following steps as shown in FIG. 9:

[0351] S901, the first device sends a fourth message to a shared capability library; correspondingly, the shared capability library receives the fourth message from the first device.

[0352] In the embodiments of the present application, the fourth message comprises the first subtask, and the fourth message is used to request the fourth capability information corresponding to the first subtask, or the fourth message is used to indicate to obtain the fourth capability information corresponding to the first subtask.

[0353] S902, the shared capability library determines the fourth capability information.

[0354] In the embodiments of the present application, the determination of the fourth capability information in S902 can be referred to the method for determining the third capability information in S702, and is not described herein.

[0355] S903, the shared capability library sends the fourth capability information to the first device; correspondingly, the first device receives the fourth capability information from the shared capability library.

[0356] It should be understood that in the case that the information of the fourth capability sent by the shared capability library to the first device includes the calling interface of the fourth capability or the source code of the fourth capability, the shared capability library can deduct the execution cost of the fourth capability from the account corresponding to the first device according to the execution cost of the fourth capability.

[0357] Through the method shown in FIG. 9, the first device can quickly determine the information of the fourth capability, facilitating subsequent determination of the second capability corresponding to the first subtask according to the information of the fourth capability.

[0358] In a possible implementation manner, after obtaining the information of the fourth capability by the method shown in FIG. 9, the first device can obtain the identifier of the second capability corresponding to the first subtask from the information of the fourth capability, and obtain the second capability corresponding to the first subtask from the shared capability library according to the identifier of the second capability corresponding to the first subtask, without the need to obtain the second capability corresponding to the first subtask by the method of S802 to S805 in FIG. 8. Through the above manner, the second capability corresponding to the first subtask can be quickly and accurately determined.

[0359] In the above S801, the third message generated by the first device can include four possible implementation manners. The manner in which the first device generates the third message can refer to the manner in which the first device generates the first message, which will not be repeated here.

[0360] S802, the second device determines the identifier of the second capability corresponding to the first subtask.

[0361] In the embodiments of the present application, the second device can determine the identifier of the second capability by any one of the following implementation manners F1, F2, F3, F4 or F5.

[0362] Implementation manner F1, in the case that the third message includes the first subtask, the second device can input the first subtask to the large language model, and output the identifier of the second capability by the large language model.

[0363] Generally, after obtaining the first subtask, the large language model can search for the capability in the capability library according to the first subtask, and then determine the identifier of the second capability corresponding to the first subtask.

[0364] Implementation manner F2, in the case that the third message includes the first subtask and the second constraint condition corresponding to the first subtask, the second device can input the first subtask and the second constraint condition to the large language model, and output the identifier of the second capability by the large language model.

[0365] Generally, after obtaining the first subtask and the second constraint condition, the large language model can search for the capability in the capability library according to the first subtask and the second constraint condition, and determine the identifier of the second capability corresponding to the first subtask. The determined second capability corresponding to the first subtask satisfies the second constraint condition.

[0366] In an implementation F3, when the third message includes the information of the first subtask and the fourth capability, the second device can input the information of the first subtask and the fourth capability to the large language model, and output the identifier of the second capability by the large language model.

[0367] Generally, after obtaining the information of the first subtask and the fourth capability, the large language model can search for the information of the fourth capability according to the first subtask, and determine the identifier of the second capability corresponding to the first subtask from the information of the fourth capability.

[0368] In an implementation F4, when the third message includes the first subtask, the second constraint condition corresponding to the first subtask, and the information of the fourth capability, the second device can input the first subtask, the second constraint condition corresponding to the first subtask, and the information of the fourth capability to the large language model, and output the identifier of the second capability by the large language model.

[0369] Generally, after obtaining the first subtask, the second constraint condition, and the information of the fourth capability, the large language model can search for the information of the fourth capability according to the first subtask and the second constraint condition, and determine the identifier of the second capability corresponding to the first subtask from the information of the fourth capability. The determined second capability corresponding to the first subtask satisfies the second constraint condition.

[0370] In an implementation F5, when the third message includes the first subtask, the second constraint condition, the information of the fourth capability, the target task, the subtask before the first subtask, and the execution result of the subtask before the first subtask, the second device can input the above information in the third message to the large language model, and output the identifier of the second capability by the large language model.

[0371] Generally, after obtaining the first subtask, the second constraint condition, the information of the fourth capability, the target task, the subtask before the first subtask, and the execution result of the subtask before the first subtask, the large language model can search for the information of the fourth capability according to the first subtask, the second constraint condition, the target task, the subtask before the first subtask, and the execution result of the subtask before the first subtask, and determine the identifier of the second capability corresponding to the first subtask from the information of the fourth capability. The determined second capability corresponding to the first subtask satisfies the second constraint condition.

[0372] If the large language model cannot determine the identifier of the second capability corresponding to the first subtask through the above five implementation manners, the large language model can further generate the second capability corresponding to the first subtask, and send the second capability corresponding to the first subtask to the first device for the first device to execute the first subtask using the second capability. In order to facilitate the description hereinafter, the following will be explained and described by taking the large language model which can determine the identifier of the second capability corresponding to the first subtask as an example.

[0373] S803, the second device sends the identifier of the second capability corresponding to the first subtask to the first device; correspondingly, the first device receives the identifier of the second capability corresponding to the first subtask from the second device.

[0374] In the embodiment of the present application, after the first device obtains the identifier of the second capability corresponding to the first subtask through the steps in S803, the first device can obtain the second capability corresponding to the first subtask through any one of the following implementation manners G1 or G2.

[0375] Implementation manner G1, the first device obtains the second capability corresponding to the first subtask from the information of the fourth capability according to the identifier of the second capability corresponding to the first subtask.

[0376] In the embodiment of the present application, in the case that the information of the fourth capability includes the calling interface of the fourth capability or the source code of the fourth capability, if the first device determines that the information of the fourth capability includes the identifier of the second capability corresponding to the first subtask, the first device can directly take the calling interface of the fourth capability or the source code of the fourth capability in the fourth capability information as the second capability corresponding to the first subtask.

[0377] Through the above implementation manner G1, the first device can obtain the second capability for executing the first subtask from the information of the fourth capability, without accessing the capability library again, which is beneficial to improve the efficiency of obtaining the second capability.

[0378] Implementation manner G2 includes the following steps as shown in S804 and S805:

[0379] S804, the first device sends a fifth message to the shared capability library; correspondingly, the shared capability library receives the fifth message from the first device.

[0380] In the embodiment of the present application, the fifth message can include the identifier of the second capability corresponding to the first subtask, and the fifth message is used to request the second capability corresponding to the first subtask, or the fifth message is used to indicate obtaining the second capability corresponding to the first subtask.

[0381] Optionally, the fifth message can further include the identifier of the first device.

[0382] Optionally, the fifth message can further include authorization information of the second capability corresponding to the first subtask, where the authorization information of the second capability corresponding to the first subtask is used to indicate authorization of the first device to acquire the second capability corresponding to the first subtask. The authorization information can be in the form of a key, a token, etc., which is not limited herein.

[0383] In the embodiments of the present application, the first device can acquire the authorization information of the second capability corresponding to the first subtask through the following steps as shown in FIG. 10.

[0384] S1001, the first device sends a sixth message to the third device; correspondingly, the third device receives the sixth message from the first device.

[0385] In the embodiments of the present application, the sixth message can include an identifier of the second capability corresponding to the first subtask, the sixth message is used to request authorization information of the second capability corresponding to the first subtask, or the sixth message is used to indicate acquisition of the authorization information of the second capability corresponding to the first subtask.

[0386] It should be understood that after the first device acquires the identifier of the second capability corresponding to the first subtask through S803, the first device can perform the steps shown in FIG. 10 to acquire the authorization information of the second capability corresponding to the first subtask.

[0387] Optionally, the sixth message can further include an identifier of the first device.

[0388] In the embodiments of the present application, the third device is a device providing authorization information of a capability.

[0389] The third device can be a CP, the CP is a capability provider, and the CP can store a capability that can be provided by the CP, information of the capability, authorization information of the capability, an authorization range of the capability, a permission range of the capability, etc., which is not limited herein. The authorization range of the capability includes an identifier of the first device that can use the capability; the permission range of the capability includes whether the source code of the capability can be acquired and / or whether the source code of the capability can be modified / inherited / derived, etc.

[0390] The third device can be an AAC (authentication authorization center), and the AAC can provide authorization information of a capability through a shared capability library or can itself provide authorization information of a capability. In the case that the AAC itself provides authorization information of a capability, the AAC can store authorization information of a capability, an authorization range of the capability, a permission range of the capability, etc., which is not limited herein. The authorization information of the capability, the authorization range of the capability, and the permission range of the capability in the AAC can come from the CP or from the shared capability library.

[0391] S1002, the third device determines whether to send the authorization information of the second capability corresponding to the first subtask to the first device according to the identifier of the second capability corresponding to the first subtask.

[0392] In the embodiments of the present application, the third device can determine whether to send the authorization information of the second capability corresponding to the first subtask to the first device by different implementation manners in different cases that the third device is a CP or an AAC.

[0393] In the case that the third device is a CP, the third device can determine whether to send the authorization information of the second capability corresponding to the first subtask to the first device by any of the following implementation manners K1 or K2.

[0394] Implementation manner K1, the CP determines to send the authorization information of the second capability corresponding to the first subtask to the first device after obtaining the identifier of the second capability corresponding to the first subtask.

[0395] In the above implementation manner K1, any first device can obtain the authorization information of the second capability corresponding to the first subtask from the CP, which is beneficial to the extensive use and sharing of the capability.

[0396] Implementation manner K2, the CP can first determine the authorization range of the second capability corresponding to the first subtask according to the identifier of the second capability corresponding to the first subtask, and if the identifier of the first device is within the authorization range of the second capability corresponding to the first subtask, it is determined to send the authorization information of the second capability corresponding to the first subtask to the first device.

[0397] In the above implementation manner K2, the CP determines whether to send the authorization information of the second capability corresponding to the first subtask to the first device according to whether the first device is within the authorization range of the second capability corresponding to the first subtask, which can effectively ensure the reasonable issuance of the authorization information.

[0398] After obtaining the authorization information of the second capability corresponding to the first subtask by the implementation manner K1 or the implementation manner K2, the CP can further determine the authority range of the second capability corresponding to the first subtask according to the identifier of the second capability corresponding to the first subtask, and if the identifier of the first device is within the authority range of the second capability corresponding to the first subtask, it indicates that the first device can obtain the source code of the second capability corresponding to the first subtask, and / or the first device can modify / inherit / derive the source code of the second capability corresponding to the first subtask, etc., and the above information is taken as the authority information of the second capability corresponding to the first subtask.

[0399] In the case that the third device is an AAC, the third device can determine whether to send the authorization information of the second capability corresponding to the first subtask to the first device by any of the following implementation manners L1, L2 or L3.

[0400] In the implementation L1, the AAC determines to send the authorization information of the second capability corresponding to the first subtask to the first device after obtaining the identification of the second capability corresponding to the first subtask.

[0401] In the implementation L1, the authorization information of the second capability corresponding to the first subtask sent by the AAC to the first device can come from the AAC, that is, the AAC can obtain the authorization information of the second capability corresponding to the first subtask from the authorization information of the capability stored by the AAC. The authorization information of the second capability corresponding to the first subtask sent by the AAC to the first device can also come from the shared capability library, that is, the AAC sends the identification of the second capability corresponding to the first subtask to the shared capability library, and the shared capability library can obtain the authorization information of the second capability corresponding to the first subtask from the shared capability library according to the identification of the second capability corresponding to the first subtask, and send the authorization information of the second capability corresponding to the first subtask to the AAC.

[0402] In the implementation L1, any first device can obtain the authorization information of the second capability corresponding to the first subtask from the AAC, which is conducive to the extensive use and sharing of the capability.

[0403] In the implementation L2, the AAC can first determine the authorization range of the second capability corresponding to the first subtask according to the identification of the second capability corresponding to the first subtask, and if the identification of the first device is within the authorization range of the second capability corresponding to the first subtask, the AAC determines to send the authorization information of the second capability corresponding to the first subtask to the first device.

[0404] In the implementation L2, the authorization information of the second capability corresponding to the first subtask sent by the AAC to the first device can come from the AAC or the shared capability library, which will not be repeated here.

[0405] In the implementation L2, the AAC determines whether to send the authorization information of the second capability corresponding to the first subtask to the first device according to whether the first device is within the authorization range of the second capability corresponding to the first subtask, which can effectively ensure the reasonable issuance of the authorization information.

[0406] In the implementation L3, the AAC can send the identification of the second capability corresponding to the first subtask and the identification of the first device to the shared capability library, the shared capability library determines the authorization range of the second capability corresponding to the first subtask according to the identification of the second capability corresponding to the first subtask, and if the identification of the first device is within the authorization range of the second capability corresponding to the first subtask, the shared capability library determines to send the authorization information of the second capability corresponding to the first subtask to the first device, and sends the authorization information of the second capability corresponding to the first subtask to the AAC, and the AAC sends the authorization information of the second capability corresponding to the first subtask to the first device.

[0407] In the above embodiment L3, the authorization information of the second capability corresponding to the first subtask sent by the AAC to the first device can come from the shared capability library.

[0408] In the above embodiment L3, the shared capability library determines whether to send the authorization information of the second capability corresponding to the first subtask to the first device according to whether the first device is within the authorized range of the second capability corresponding to the first subtask, which can effectively ensure that the authorization information is reasonably issued.

[0409] After obtaining the authorization information of the second capability corresponding to the first subtask through the embodiment L1 or the embodiment L2 or the embodiment L3, the AAC can further determine the permission information of the second capability corresponding to the first subtask. The way in which the AAC determines the permission information of the second capability corresponding to the first subtask can refer to the way in which the CP determines the permission information of the second capability corresponding to the first subtask, which will not be repeated here. S1003, in the case where the third device determines to send the authorization information of the second capability corresponding to the first subtask to the first device, the third device sends the authorization information of the second capability corresponding to the first subtask to the first device.

[0410] Through the above method shown in FIG. 10, the first device can quickly obtain the authorization information of the second capability corresponding to the first subtask, which is conducive to improving the efficiency of obtaining the authorization information of the second capability corresponding to the first subtask.

[0411] S805, the shared capability library sends the second capability corresponding to the first subtask to the first device according to the identifier of the second capability corresponding to the first subtask.

[0412] In the embodiments of the present application, the shared capability library can determine the second capability corresponding to the first subtask through any one of the following embodiments N1, N2 or N3.

[0413] In the embodiment N1, in the case where the fifth message includes the identifier of the second capability corresponding to the first subtask, if the shared capability library does not need to perform authorization information review, the shared capability library can obtain the second capability corresponding to the first subtask according to the identifier, and send the second capability corresponding to the first subtask to the first device; if the shared capability library needs to perform authorization information review, since the authorization information of the second capability corresponding to the first subtask is not included in the fifth message, the shared capability library refuses to send the second capability corresponding to the first subtask to the first device.

[0414] In the embodiment N2, in the case where the fifth message includes the identifier and the authorization information of the second capability corresponding to the first subtask, the shared capability library confirms that the authorization information of the second capability corresponding to the first subtask is included in the fifth message, the shared capability library obtains the second capability corresponding to the first subtask according to the identifier, and sends the second capability corresponding to the first subtask to the first device.

[0415] Optionally, before performing the implementation manner N1 or the implementation manner N2, the shared capability library can verify the authorization information of the second capability corresponding to the first subtask, and in a case where it is determined that the authorization information is valid authorization information, the shared capability library performs the implementation manner N1 or the implementation manner N2; in a case where it is determined that the authorization information is invalid authorization information, the shared capability library rejects to send the second capability corresponding to the first subtask to the first device.

[0416] In a case where the fifth message includes the identifier of the second capability corresponding to the first subtask and the identifier of the first device, the shared capability library determines the authorization range of the second capability corresponding to the first subtask according to the identifier of the second capability corresponding to the first subtask, and if the identifier of the first device is within the authorization range of the second capability corresponding to the first subtask, the shared capability library can acquire the second capability corresponding to the first subtask according to the identifier and send the second capability corresponding to the first subtask to the first device.

[0417] In addition, in a case where the fifth message includes the permission information of the second capability corresponding to the first subtask, the second capability corresponding to the first subtask finally determined by the shared capability library needs to satisfy the content of the permission information.

[0418] For example, the permission information of the second capability corresponding to the first subtask includes that the source code of the second capability corresponding to the first subtask can be acquired and modified, and the source code of the second capability corresponding to the first subtask cannot be inherited and derived, and the permission mechanism of the source code of the second capability corresponding to the first subtask sent by the shared capability library to the first device is to allow use, allow modification, and not allow inheritance and derivation. It should be understood that the permission mechanism of the source code of the second capability corresponding to the first subtask is the mechanism set in the source code of the second capability corresponding to the first subtask.

[0419] In a possible implementation manner, when the shared capability library sends the second capability corresponding to the first subtask to the first device, the shared capability library can also acquire the execution cost of the second capability corresponding to the first subtask from the information of the second capability corresponding to the first subtask, and deduct the execution cost of the second capability corresponding to the first subtask from the account corresponding to the first device. The shared capability library can also add the deducted execution cost of the second capability corresponding to the first subtask to the account corresponding to the first device that provides the second capability corresponding to the first subtask.

[0420] Through the above implementation manner G2, the first device can acquire the second capability for executing the first subtask from the shared capability library.

[0421] Through the above embodiment B2, in a case where the first device cannot directly obtain the first capability corresponding to the target task, the first device can split the target task into at least one subtask, and obtain the second capability corresponding to the subtask to achieve the purpose of obtaining the first capability, so that the above process can achieve the purpose of obtaining the first capability. S303, the first device uses the first capability to perform the target task.

[0422] In a case where the first capability is in an interface sharing mode, the first device invokes the invocation interface of the first capability to achieve the purpose of using the first capability to perform the target task, wherein the invocation interface of the first capability can be provided by the shared capability library or the CP.

[0423] In a case where the first capability is in a source code sharing mode, after the first device obtains the source code of the first capability, the first device performs the target task by using the source code of the first capability.

[0424] Through the method shown in FIG. 3, different first devices can obtain the first capability corresponding to the target task from the shared capability library when processing similar target tasks, and use the first capability to perform the target task to obtain an execution result, instead of processing the target task and generating a capability for performing the target task respectively. The above process can reduce unnecessary homogenization operations, and is beneficial to reduce the time spent in the task processing process and improve the task processing efficiency of the first device.

[0425] After the first device performs the target task by using the first capability through the method shown in FIG. 3, the first device can add the capability generated in the method shown in FIG. 3 to the shared capability library and / or the local capability library corresponding to the first device. Taking an example in which the first device adds the generated capability to the shared capability library, refer to the following steps shown in FIG. 11.

[0426] Optionally, the method of FIG. 11 includes S1101.

[0427] S1101, the first device determines a fifth capability according to the second capability corresponding to each of the plurality of subtasks in the at least one subtask. The fifth capability is used to implement the target task or part of the target task.

[0428] In the embodiments of the present application, the first device can determine the fifth capability through any one of the following embodiments P1 or P2.

[0429] Embodiment P1, the first device can generate the fifth capability according to the capability corresponding to each of the plurality of subtasks in the at least one subtask. The generated fifth capability can implement the plurality of subtasks.

[0430] Through the above embodiment P1, the first device can quickly generate the fifth capability, which is beneficial to improve the efficiency of generating new capabilities.

[0431] Embodiment P2 includes the following steps as shown in FIG. 12.

[0432] S1201, the first device sends an eighth message to the second device; correspondingly, the second device receives the eighth message from the first device.

[0433] In the embodiment of the present application, the eighth message can include a second capability corresponding to each of a plurality of subtasks in the at least one subtask, and the eighth message is used to indicate that a capability is generated according to the second capability corresponding to each of the plurality of subtasks in the at least one subtask.

[0434] S1202, the second device sends a fifth capability to the first device; correspondingly, the first device receives the fifth capability from the second device.

[0435] In the embodiment of the present application, after obtaining the second capability corresponding to each of the plurality of subtasks in the eighth message, the second device inputs the second capability corresponding to each of the plurality of subtasks to the large language model, outputs the fifth capability by the large language model, and sends the fifth capability to the first device.

[0436] In the above embodiment P2, the fifth capability is generated by the large language model in the second device instead of the first device, which can reduce the power consumption of the first device. In addition, the fifth capability generated by the large language model is more complete and reasonable.

[0437] S1102, the first device sends a seventh message to the shared capability library.

[0438] In the embodiment of the present application, the seventh message can include any of the following embodiments Q1 or Q2.

[0439] Embodiment Q1, the seventh message includes a second capability corresponding to any subtask in the at least one subtask, and the seventh message is used to indicate that the second capability corresponding to any subtask is added to the shared capability library.

[0440] Optionally, in the case that the seventh message includes the second capability corresponding to the first subtask, the seventh message can further include at least one of the following:

[0441] 1) an identifier of the first device;

[0442] 2) description information of the second capability corresponding to the first subtask, the description information of the second capability corresponding to the first subtask being used to describe the function of the second capability corresponding to the first subtask;

[0443] 3) a sharing mode of the second capability corresponding to the first subtask, the sharing mode of the second capability corresponding to the first subtask being used to indicate the presentation mode of the second capability corresponding to the first subtask;

[0444] 4) parameter information of the second capability corresponding to the first subtask, the parameter information of the second capability corresponding to the first subtask being used to indicate input parameters required for using the second capability corresponding to the first subtask and / or to indicate output parameters of the second capability;

[0445] 5) environment dependency information of the second capability corresponding to the first subtask, the environment dependency information of the second capability corresponding to the first subtask being used to indicate environment information on which the second capability corresponding to the first subtask depends;

[0446] 6) execution time of the second capability corresponding to the first subtask, the execution time of the second capability corresponding to the first subtask being used to indicate time required for executing the second capability corresponding to the first subtask;

[0447] 7) execution cost of the second capability corresponding to the first subtask, the execution cost of the second capability corresponding to the first subtask being used to indicate cost required for using the second capability corresponding to the first subtask;

[0448] 8) calling interface of the second capability corresponding to the first subtask, the calling interface of the second capability corresponding to the first subtask being used to implement a function of the second capability corresponding to the first subtask;

[0449] 9) source code of the second capability corresponding to the first subtask, the source code of the second capability corresponding to the first subtask being used to implement a function of the second capability corresponding to the first subtask;

[0450] 10) source code permission policy of the second capability corresponding to the first subtask, the source code permission policy of the second capability corresponding to the first subtask being used to indicate whether the source code of the second capability corresponding to the first subtask is allowed to be modified / inherited / derived and / or whether other capabilities generated based on the source code of the second capability corresponding to the first subtask need to be added to the shared capability library.

[0451] It should be understood that the above information of the second capability corresponding to the first subtask can be set by a user, for example, the user can set the execution cost of the second capability corresponding to the first subtask to be 100 points; the above information of the second capability corresponding to the first subtask can also be obtained from a capability library.

[0452] It should be understood that detailed descriptions of the above information of the second capability corresponding to the first subtask can refer to the detailed descriptions of the information of the third capability in the foregoing, and will not be described here.

[0453] Optionally, the seventh message can further include an encoding vector of the information of the second capability corresponding to the first subtask, wherein the encoding vector of the information of the second capability corresponding to the first subtask can be generated by the first device or can be generated by a large language model in the second device, which is not limited here.

[0454] In an implementation Q2, the seventh message comprises a fifth capability, and the seventh message is used to indicate that the fifth capability is added to the shared capability library.

[0455] Optionally, the seventh message further comprises at least one of:

[0456] 1) an identity of the first device;

[0457] 2) description information of the fifth capability, the description information of the fifth capability being used to describe a function of the fifth capability;

[0458] 3) a sharing manner of the fifth capability, the sharing manner of the fifth capability being used to indicate a presentation manner of the fifth capability;

[0459] 4) parameter information of the fifth capability, the parameter information of the fifth capability being used to indicate an input parameter required for using the fifth capability and / or to indicate an output parameter of the fifth capability;

[0460] 5) environment dependency information of the fifth capability, the environment dependency information of the fifth capability being used to indicate environment information on which the fifth capability depends;

[0461] 6) execution time of the fifth capability, the execution time of the fifth capability being used to indicate a time required for executing the fifth capability;

[0462] 7) execution cost of the fifth capability, the execution cost of the fifth capability being used to indicate a cost required for using the fifth capability;

[0463] 8) a calling interface of the fifth capability, the calling interface of the fifth capability being used to implement the function of the fifth capability;

[0464] 9) source code of the fifth capability, the source code of the fifth capability being used to implement the function of the fifth capability;

[0465] 10) source code permission policy of the fifth capability, the source code permission policy of the fifth capability being used to indicate whether the source code of the fifth capability allows modification / inheritance / derivation and / or whether other capabilities generated based on the source code of the fifth capability need to be added to the shared capability library.

[0466] It should be understood that the above information of the fifth capability can be set by a user, or can be generated by the first device or the second device when the fifth capability is generated, which is not limited herein.

[0467] Further, since the fifth capability is generated according to the capabilities corresponding to the plurality of sub-tasks in the at least one sub-task respectively, the source code permission policy of the fifth capability is according to the source code permission policies of the capabilities corresponding to the plurality of sub-tasks respectively. For example, the capabilities corresponding to the 2 sub-tasks for generating the fifth capability are capability 1 and capability 2 respectively, the source code permission policy of capability 1 indicates that the source code of capability 1 is not allowed to be modified, and the source code permission policy of capability 2 indicates that the source code of capability 2 is allowed to be modified, then the source code permission policy of the fifth capability includes that the source code of the fifth capability is not allowed to be modified.

[0468] It should be understood that the detailed description of the information about the fifth capability above can refer to the detailed description of the information about the third capability above, which will not be repeated here.

[0469] Optionally, the seventh message can further include an encoding vector of the information of the fifth capability, wherein the encoding vector of the information of the fifth capability can be generated by the first device or by the large language model in the second device, which is not limited here.

[0470] Optionally, before performing S1102, the first device can verify the second capability or the fifth capability corresponding to the first sub-task, and after the second capability or the fifth capability corresponding to the first sub-task is verified, the first device generates and sends the seventh message according to the second capability or the fifth capability corresponding to the first sub-task. Alternatively, the first device can send the second capability or the fifth capability corresponding to the first sub-task to the second device, and the second device verifies the second capability or the fifth capability corresponding to the first sub-task. After the second device verifies the second capability or the fifth capability corresponding to the first sub-task, the second device sends a verification pass message to the first device, and after the first device receives the verification pass message, the first device generates and sends the seventh message according to the second capability or the fifth capability corresponding to the first sub-task.

[0471] The following introduces how the first device verifies the fifth capability, and the verification of the second capability corresponding to the first sub-task by the first device can refer to the verification of the fifth capability by the first device, which will not be repeated here. The verification of the second capability or the fifth capability corresponding to the first sub-task by the second device can refer to the verification of the second capability or the fifth capability corresponding to the first sub-task by the first device, which will not be repeated here.

[0472] In a possible implementation manner, the verification of the fifth capability can include at least one of the following:

[0473] 1) Verification of the execution time of the fifth capability by the first device.

[0474] In the embodiments of the present application, the first device can determine whether the execution time of the fifth capability is less than a first threshold, and if the execution time of the fifth capability is less than the first threshold, the verification is passed; otherwise, the verification is not passed.

[0475] 2) The first device verifies the dependency environment of the fifth capability.

[0476] In the embodiment of the present application, the first device can determine whether the number of the dependency environment of the fifth capability is less than the second threshold value. If the number of the dependency environment of the fifth capability is less than the second threshold value, the verification is passed; otherwise, the verification is failed.

[0477] The first device can further determine whether the dependency environment of the fifth capability includes a preset environment. If the dependency environment of the fifth capability includes the preset environment, the verification is failed; otherwise, the verification is passed. The preset environment can be a type of operating system, CPU, network card, etc. which is pre-set, and is not limited herein.

[0478] 3) The first device verifies the code corresponding to the fifth capability.

[0479] In the embodiment of the present application, the first device can determine whether the number of lines of the code corresponding to the fifth capability is less than the third threshold value. If the number of lines of the code corresponding to the fifth capability is less than the third threshold value, the verification is passed; otherwise, the verification is failed.

[0480] The first device can further determine whether there is private data in the code corresponding to the fifth capability. If there is private data in the code corresponding to the fifth capability, the verification is failed; otherwise, the verification is passed. The private data can include, but is not limited to, user private information, etc.

[0481] The first device can further determine whether there is a high-risk interface in the code corresponding to the fifth capability. If there is a high-risk interface in the code corresponding to the fifth capability, the verification is failed; otherwise, the verification is passed. The high-risk interface can include, but is not limited to, a high-risk system interface, a user interface, etc.

[0482] S1103, the shared capability library adds the second capability or the fifth capability corresponding to any of the at least one subtask to the shared capability library.

[0483] In the embodiment of the present application, before the first device performs S1103, in the case that the seventh message includes the fifth capability, the shared capability library can perform similarity checking on the fifth capability.

[0484] In a possible implementation manner, the shared capability library can calculate the similarity of the fifth capability and the capabilities in the shared capability library. If there is no similarity greater than a second similarity threshold value, the shared capability library can add the fifth capability to the shared capability library; if there is a similarity greater than the second similarity threshold value, the shared capability library can perform capability processing on the fifth capability.

[0485] The capability processing includes any of the following:

[0486] 1) the shared capability library deletes the capability similar to the fifth capability, and adds the fifth capability to the shared capability library;

[0487] 2) the shared capability library marks the fifth capability and the capability similar to the fifth capability, and adds the fifth capability to the shared capability library;

[0488] 3) the shared capability library rejects to add the fifth capability.

[0489] Optionally, after the shared capability library successfully stores the fifth capability, the shared capability library can also issue a reward corresponding to the fifth capability to the account of the first device. The amount of the reward of the fifth capability issued by the shared capability library can be determined by the shared capability library or by the first device, which is not described herein.

[0490] In addition, the shared capability library can also generate an identifier of the fifth capability, and take the identifier of the fifth capability, the description information of the fifth capability, the sharing manner of the fifth capability, the parameter information of the fifth capability, the environment dependency information of the fifth capability, the execution time of the fifth capability, the execution cost of the fifth capability, the calling interface of the fifth capability, the source code of the fifth capability, and the source code permission policy of the fifth capability as the information of the fifth capability. The shared capability library can also store the information of the fifth capability.

[0491] It should be understood that the identifier of the fifth capability can be generated by the shared capability library, or generated by the first device and sent to the shared capability library, which is not limited herein.

[0492] In addition, the shared capability library can also encode the information of the fifth capability to obtain an encoding vector of the information of the fifth capability, and the shared capability library can also store the encoding vector of the information of the fifth capability.

[0493] It should be understood that in the case that the seventh message includes the second capability corresponding to any of the at least one subtask, the processing of the shared capability library on the second capability corresponding to any of the at least one subtask can refer to the processing of the fifth capability described above, which is not described herein.

[0494] Optionally, the method of FIG. 11 can further include S1104.

[0495] S1104, the shared capability library sends a response message to the first device; correspondingly, the first device receives the response message from the shared capability library.

[0496] In the embodiments of the present application, the response message is used to indicate that the second capability or the fifth capability corresponding to any of the at least one subtask is successfully added to the shared capability library.

[0497] Optionally, the response message can further include a subscription address corresponding to the fifth capability, and the first device can access the subscription address corresponding to the fifth capability to obtain the number of times of use of the fifth capability, the identity of the first device using the fifth capability, whether the source code of the fifth capability is modified / inherited / derived, the identity of other capabilities derived based on the source code of the fifth capability, and the like, which are not limited herein.

[0498] Optionally, the shared capability library can further receive a subscription request message from the first device, where the subscription request message can include the identity of the fifth capability, and can further include at least one of the following: request information of the number of times of use of the fifth capability, request information of the identity of the first device using the fifth capability, request information of whether the source code of the fifth capability is modified / inherited / derived, and request information of the identity of other capabilities derived based on the source code of the fifth capability. The shared capability library sends a subscription response message to the first device, where the subscription response message can include a subscription address corresponding to the fifth capability, and the first device can access the subscription address corresponding to the fifth capability to obtain at least one of the following: the number of times of use of the fifth capability, the identity of the first device using the fifth capability, whether the source code of the fifth capability is modified / inherited / derived, and the identity of other capabilities derived based on the source code of the fifth capability.

[0499] Optionally, the shared capability library can further receive a capability modification message from the first device, where the capability modification message includes the identity of the fifth capability and modification information. The modification information can include modifications to at least one of the following: description information of the fifth capability, sharing mode of the fifth capability, parameter information of the fifth capability, environment dependency information of the fifth capability, execution time of the fifth capability, execution cost of the fifth capability, calling interface of the fifth capability, source code of the fifth capability, and source code permission policy of the fifth capability. The shared capability library modifies the fifth capability or information of the fifth capability according to the modification information in the capability modification message.

[0500] Optionally, the shared capability library can further receive a capability cancellation message from the first device, where the capability cancellation message includes the identity of the fifth capability, and the capability cancellation message is used to cancel the fifth capability. The shared capability library deletes the fifth capability and information of the fifth capability from the shared capability library after obtaining the capability cancellation message.

[0501] It should be understood that the subscription request message, the capability modification message, and the capability cancellation message corresponding to the second capability of any subtask can refer to the subscription request message, the capability modification message, and the capability cancellation message of the fifth capability, which are not described herein.

[0502] Through the above method shown in FIG. 11, the first device can add the generated capability to the shared capability library, so that other first devices can obtain and use the capability from the shared capability library, thereby avoiding unnecessary homogenization operations.

[0503] In the embodiments of the present application, the shared capability library can also periodically or non-periodically optimize the capabilities in the shared capability library. The shared capability library can also periodically or non-periodically send the capabilities in the shared capability library to the second device, and the large language model in the second device can optimize the capabilities in the shared capability library. For the convenience of the following description, the following will be described in detail by taking the shared capability library for capability optimization as an example. The large language model in the second device for capability optimization can refer to the shared capability library for capability optimization, and details are not described herein.

[0504] In a possible implementation, the capability optimization trigger condition corresponding to the non-periodic capability optimization of the shared capability library can include at least one of the following: the remaining storage space of the shared capability library is less than a fourth threshold value, or the number of capabilities in the shared capability library is greater than a fifth threshold value, or the sum of the code lines corresponding to each capability in the shared capability library is greater than a sixth threshold value, etc., which is not limited herein.

[0505] The capability optimization of the capabilities in the shared capability library includes at least one of the following: capability deletion, capability merging, capability replacement, and capability reconstruction. The capability optimization is explained in detail as follows.

[0506] 1) Capability deletion

[0507] For any capability in any shared capability library, if the time when the capability is added to the shared capability library exceeds a seventh threshold value, or the number of times the capability is used is less than an eighth threshold value, or the probability of executing the capability is greater than a ninth threshold value, the shared capability library can delete the capability.

[0508] 2) Capability merging

[0509] For any two capabilities in any shared capability library, if the similarity of the functions of the above two capabilities is greater than a ninth threshold value, or the similarity of the dependent environments of the above two capabilities is greater than a tenth threshold value, or the above two capabilities are added by the same first device, the above two capabilities can be merged.

[0510] 3) Capability replacement

[0511] For any two capabilities in any shared capability library, if the similarity of the functions of the above two capabilities is greater than a ninth threshold value, or the similarity of the dependent environments of the above two capabilities is greater than a tenth threshold value, or the above two capabilities are added by the same first device, one of the capabilities can be used to replace the other capability.

[0512] 4) Capability reconstruction

[0513] For any capability in any shared capability library, if there is a logical error in the source code corresponding to the capability, or there is a branch in the source code corresponding to the capability with a time consumption greater than an eleventh threshold value, the source code of the capability can be rewritten.

[0514] For any capability in any shared capability library, if there is a part of code in the source code corresponding to the capability that can be replaced by other capabilities, the replaceable capability can be used to replace this part of code.

[0515] It should be understood that in the case where the first device acquires the subscription address corresponding to the fifth capability, if the fifth capability is deleted, merged, replaced, or reconstructed, the first device will acquire information about the deletion, merging, replacement, or reconstruction of the fifth capability through the subscription address.

[0516] In the above method, by optimizing the capabilities in the shared capability library, the quality of the capabilities in the shared capability library can be continuously improved, and the execution efficiency of the capabilities can be ensured.

[0517] In the above embodiments of the present application, the methods provided by the embodiments of the present application are introduced from the perspective of interaction between devices. To implement the functions in the above methods provided by the embodiments of the present application, the first device can include hardware structures and / or software modules to implement the above functions in the form of hardware structures, software modules, or hardware structures plus software modules. Whether a certain function in the above functions is executed in the form of hardware structures, software modules, or hardware structures plus software modules depends on the specific application of the technical solution and the design constraint conditions.

[0518] The division of modules in the embodiments of the present application is illustrative, and is only a logical functional division. In actual implementation, another division manner can be used. In addition, the functional modules in each embodiment of the present application can be integrated in one processor, or can be physically separated, or two or more modules can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software functional module.

[0519] In the following, the communication device provided by the embodiments of the present application will be described in detail in combination with FIGS. 13 to 14. It should be understood that the description of the device embodiments corresponds to the description of the method embodiments, and therefore, the content not described in detail can be referred to the method embodiments described above, and will not be described here again for brevity.

[0520] The same as the above idea, as shown in FIG. 13, the embodiment of the present application further provides a communication device 1300 for implementing the functions of the first device or the shared capability library in the above method. For example, the communication device can be a software module or a chip system. In the embodiment of the present application, the chip system can be composed of a chip, or can contain a chip and other discrete devices. The communication device 1300 can include a communication unit 1301 and a processing unit 1302.

[0521] In the embodiment of the present application, the communication unit and the processing unit of the communication device 1300 are respectively used to execute the sending and receiving steps of the first device or the shared capability library in the above method embodiment. The communication unit and the processing unit can be integrated together as one unit, or can be two independent units. In one implementation manner, the communication device 1300 can execute the following functions:

[0522] The communication unit 1301 is used to acquire a target task.

[0523] The processing unit 1302 is used to acquire a first capability from the shared capability library according to the target task, wherein the shared capability library is used to store capabilities, the capabilities in the shared capability library include the first capability, and the shared capability library supports multiple first devices sharing the capabilities in the shared capability library.

[0524] The processing unit 1302 is further used to execute the target task by using the first capability.

[0525] In one possible design, the processing unit 1302 is specifically used to determine at least one subtask corresponding to the target task, and the communication unit 1301 is specifically used to acquire second capabilities respectively corresponding to the at least one subtask from the shared capability library, wherein the first capability includes the second capabilities respectively corresponding to the at least one subtask.

[0526] In one possible design, the processing unit 1302 is specifically used to send a first message to a second device, the first message includes the target task, and the first message is used to request at least one subtask corresponding to the target task; and the processing unit 1302 is further used to receive the at least one subtask from the second device.

[0527] In one possible design, the first message further includes information of a third capability, and the information of the third capability is used to determine the at least one subtask.

[0528] The communication unit 1301 is further used to send a second message to a capability library, the second message includes the target task, and the second message is used to request information of a capability corresponding to the target task; the capability library includes a local capability library corresponding to the first device and / or the shared capability library; the local capability library corresponding to the first device is used to store capabilities, and the local capability library corresponding to the first device supports the first device to acquire the capabilities in the local capability library.

[0529] The communication unit 1301 is further configured to receive the information of the third capability from the capability library.

[0530] In a possible design, the second message further includes a first constraint condition corresponding to the target task; and the third capability satisfies the first constraint condition.

[0531] In a possible design, the at least one subtask includes a first subtask; and the communication unit 1301 is specifically configured to: send, to the second device, a third message, where the third message includes the first subtask, and the third message is used to request an identification of a second capability corresponding to the first subtask; receive, from the second device, the identification of the second capability corresponding to the first subtask; and obtain, from the shared capability library, the second capability corresponding to the first subtask according to the identification of the second capability corresponding to the first subtask.

[0532] In a possible design, the third message further includes information of a fourth capability corresponding to the first subtask, and the information of the fourth capability is used to determine the identification of the second capability corresponding to the first subtask; and the communication unit 1301 is further configured to: send, to the capability library, a fourth message, where the fourth message includes the first subtask, and the fourth message is used to request information of a capability corresponding to the first subtask; and receive, from the capability library, the information of the fourth capability.

[0533] In a possible design, the fourth message further includes a second constraint condition corresponding to the first subtask, and the second constraint condition is determined according to the first constraint condition corresponding to the target task; and the fourth capability satisfies the second constraint condition.

[0534] In a possible design, the communication unit 1301 is specifically configured to: send, to the shared capability library, a fifth message, where the fifth message includes the identification of the second capability corresponding to the first subtask, and the fifth message is used to request the second capability corresponding to the first subtask; and receive, from the shared capability library, the second capability corresponding to the first subtask.

[0535] In a possible design, the communication unit 1301 is specifically configured to: obtain, from the information of the fourth capability, the second capability corresponding to the first subtask according to the identification of the second capability corresponding to the first subtask, where the information of the fourth capability is from the shared capability library.

[0536] In a possible design, the communication unit 1301 is specifically configured to: obtain, from the shared capability library, the second capability corresponding to the first subtask according to the identification of the second capability corresponding to the first subtask and authorization information, where the authorization information is used to indicate that the first device is authorized to obtain the second capability corresponding to the first subtask.

[0537] In a possible design, the communication unit 1301 is further configured to: send, to the third device, a sixth message, where the sixth message includes an identifier of the second capability corresponding to the first subtask, and the sixth message is used to request authorization information of the second capability corresponding to the first subtask; and receive, from the third device, the authorization information of the second capability corresponding to the first subtask.

[0538] In a possible design, the communication unit 1301 is further configured to: send, to the shared capability library, a seventh message, where the seventh message includes the second capability corresponding to any of the at least one subtask, and the seventh message is used to indicate that the second capability corresponding to any of the at least one subtask is added to the shared capability library.

[0539] In a possible design, the communication unit 1301 is further configured to: determine a fifth capability according to the second capabilities corresponding to the plurality of subtasks of the at least one subtask respectively, where the fifth capability is used to implement the target task or is used to implement a part of the target task; and send, to the shared capability library, a seventh message, where the seventh message includes the fifth capability, and the seventh message is used to indicate that the fifth capability is added to the shared capability library.

[0540] In a possible design, the communication unit 1301 is specifically configured to: send, to the second device, an eighth message, where the eighth message includes the second capabilities corresponding to the plurality of subtasks of the at least one subtask respectively, and the eighth message is used to indicate that one capability is generated according to the second capabilities corresponding to the plurality of subtasks of the at least one subtask respectively; and receive, from the second device, the fifth capability.

[0541] In an implementation, the communication apparatus 1300 can perform the following functions:

[0542] The communication unit 1301 is configured to receive, from the first device, a task.

[0543] The communication unit 1301 is further configured to send, to the first device, a first capability corresponding to the task, where the first capability is used to execute the task.

[0544] In a possible design, the communication unit 1301 is specifically configured to: receive, from the first device, a target task; and send, to the first device, a first capability corresponding to the target task.

[0545] In a possible design, the communication unit 1301 is specifically configured to: receive, from the first device, at least one subtask corresponding to a target task; and send, to the first device, second capabilities corresponding to the at least one subtask respectively, where the first capability includes the second capabilities corresponding to the at least one subtask respectively.

[0546] In a possible design, the at least one subtask includes a first subtask; and the communication unit 1301 is specifically configured to: receive a fifth message from the first device, where the fifth message includes identification of the second capability corresponding to the first subtask, and the fifth message is used to request the second capability corresponding to the first subtask; and send, to the first device, the second capability corresponding to the first subtask.

[0547] In a possible design, the processing unit 1302 is configured to deduct, from an account corresponding to the first device, an execution cost of the second capability corresponding to the first subtask.

[0548] In a possible design, the communication unit 1301 is further configured to: receive a seventh message from the first device, where the seventh message includes the second capability corresponding to any of the at least one subtask, and the seventh message is used to indicate that the second capability corresponding to any of the at least one subtask is added to the shared capability library; and the processing unit 1302 is further configured to: add the second capability corresponding to any of the at least one subtask to the shared capability library.

[0549] In a possible design, the processing unit 1302 is further configured to: pay, to an account corresponding to the first device, a reward corresponding to the second capability corresponding to any of the at least one subtask.

[0550] In a possible design, the communication unit 1301 is further configured to: receive a seventh message from the first device, where the seventh message includes a fifth capability, and the fifth capability is generated according to the second capabilities corresponding to the plurality of subtasks in the at least one subtask, and the seventh message is used to indicate that the fifth capability is added to the shared capability library; and the processing unit 1302 is further configured to: add the fifth capability to the shared capability library.

[0551] In a possible design, the processing unit 1302 is further configured to: pay, to an account corresponding to the first device, a reward corresponding to the fifth capability.

[0552] In a possible design, the processing unit 1302 is further configured to: optimize the capabilities in the shared capability library.

[0553] The above is merely an example, and the communication unit and the processing unit in the communication device 1300 can also perform other functions, and more detailed descriptions can be referred to related descriptions in the method embodiments shown above, which are not repeated here.

[0554] The method provided in the embodiments of the present application is described from the perspective of the communication device as the execution subject. To implement the functions in the method provided in the embodiments of the present application, the communication device can include a hardware structure and / or a software module, and implement the above functions in the form of the hardware structure, the software module, or the hardware structure plus the software module. Whether a certain function in the above functions is implemented in the form of the hardware structure, the software module, or the hardware structure plus the software module depends on the specific application and design constraints of the technical solution.

[0555] For example, when implemented in hardware, the hardware implementation of the communication device can refer to FIG. 14 and the related description thereof.

[0556] Referring to FIG. 14, the communication device can include one or more processors 1402, a memory 1403, and one or more computer programs 1404, which can be connected through one or more communication buses 1401. The one or more computer programs 1404 are stored in the memory 1403 and configured to be executed by the one or more processors 1402, and the one or more computer programs 1404 include instructions that can be used to execute the method in any of the above embodiments. The one or more processors 1402 can execute the functions of the communication unit 1301 and the processing unit 1302.

[0557] The embodiments of the present application also provide a computer readable storage medium, which stores computer instructions, and when the computer instructions run on the communication device, the communication device implements the communication method in the above embodiments.

[0558] The embodiments of the present application also provide a computer program product, which, when running on a computer, causes the computer to execute the communication method in the above embodiments.

[0559] The communication device, the computer readable storage medium, the computer program product, or the chip provided in the embodiments of the present application are used to execute the corresponding method provided above, and thus the beneficial effects achieved thereby can refer to the beneficial effects of the corresponding method provided above, which will not be described herein again.

[0560] Through the description of the above embodiments, those skilled in the art can understand that, for the convenience and brevity of description, only the division of the above functional modules is taken as an example for illustration, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above.

[0561] In several embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other manners. For example, the division of the apparatus embodiments is merely an example, and the division of the modules or units can be different, for example, multiple modules or units can be combined or integrated into another apparatus, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, apparatuses or units, and can be in electrical, mechanical or other forms.

[0562] The units described as separate components may or may not be physically separate, and the components shown as units may be one physical unit or multiple physical units, i.e., may be located in one place, or may be distributed in multiple different places. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiments.

[0563] In addition, each functional unit in the various embodiments of the present application can be integrated into one processing unit, or each unit can be physically present separately, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0564] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present application essentially or the parts that make contributions to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium, includes several instructions to make a device (which can be a single chip, a chip, etc.) or a processor execute all or part of the steps of the various embodiments of the method of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0565] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A communication method characterized by comprising: The method is applied to a first device, and comprises: obtaining a target task; obtaining a first capability from a shared capability library according to the target task, wherein the shared capability library is used to store capabilities, the capabilities in the shared capability library include the first capability, and the shared capability library supports multiple first devices sharing the capabilities in the shared capability library; performing the target task by using the first capability.

2. The method of claim 1, wherein, The step of obtaining the first capability from the shared capability library according to the target task comprises: determining at least one subtask corresponding to the target task; obtaining a second capability corresponding to each of the at least one subtask from the shared capability library, wherein the first capability includes the second capability corresponding to each of the at least one subtask.

3. The method of claim 2, wherein, The step of determining the at least one subtask corresponding to the target task comprises: sending a first message to a second device, wherein the first message includes the target task, and the first message is used to request the at least one subtask corresponding to the target task; receiving the at least one subtask from the second device.

4. The method of claim 3, wherein, The first message further includes information of a third capability, wherein the information of the third capability is used to determine the at least one subtask. The method further comprises: sending a second message to a capability library, wherein the second message includes the target task, and the second message is used to request information of a capability corresponding to the target task; the capability library includes a local capability library corresponding to the first device and / or the shared capability library; the local capability library corresponding to the first device is used to store capabilities, and the local capability library corresponding to the first device supports the first device to obtain the capabilities in the local capability library; receiving the information of the third capability from the capability library.

5. The method of any one of claims 3-4, wherein, The second message further includes a first constraint condition corresponding to the target task; and the third capability satisfies the first constraint condition.

6. The method of any one of claims 2-5, wherein, The at least one subtask includes a first subtask. The step of obtaining the second capability corresponding to each of the at least one subtask from the shared capability library comprises: sending a third message to a second device, wherein the third message includes the first subtask, and the third message is used to request an identifier of a second capability corresponding to the first subtask; receiving the identifier of the second capability corresponding to the first subtask from the second device; obtaining the second capability corresponding to the first subtask from the shared capability library according to the identifier of the second capability corresponding to the first subtask.

7. The method of claim 6, wherein, The third message further includes information of a fourth capability corresponding to the first subtask, wherein the information of the fourth capability is used to determine the identifier of the second capability corresponding to the first subtask. The method further comprises: sending a fourth message to the capability library, wherein the fourth message includes the first subtask, and the fourth message is used to request information of a capability corresponding to the first subtask; receiving the information of the fourth capability from the capability library.

8. The method of claim 7, wherein, The fourth message further includes a second constraint condition corresponding to the first subtask, wherein the second constraint condition is determined according to the first constraint condition corresponding to the target task; and the fourth capability satisfies the second constraint condition.

9. The method of any one of claims 6-8, wherein, The method further comprises: sending a sixth message to a third device, the sixth message comprising an identifier of the second capability corresponding to the first subtask, the sixth message being used to request authorization information of the second capability corresponding to the first subtask; receiving the authorization information of the second capability corresponding to the first subtask from the third device.

10. The method of any one of claims 7-8, wherein, The method further comprises: sending a seventh message to the shared capability library, the seventh message comprising the second capability corresponding to any subtask of the at least one subtask, the seventh message being used to indicate that the second capability corresponding to the any subtask is added to the shared capability library.

11. The method of any one of claims 6-10, wherein, The method further comprises: determining a fifth capability according to the second capabilities corresponding to the plurality of subtasks of the at least one subtask, the fifth capability being used to implement the target task or to implement a part of the target task; 12. The method of claim 11, wherein, sending a seventh message to the shared capability library, the seventh message comprising the fifth capability, the seventh message being used to indicate that the fifth capability is added to the shared capability library. The method further comprises: sending an eighth message to the second device, the eighth message comprising the second capabilities corresponding to the plurality of subtasks of the at least one subtask, the eighth message being used to indicate that one capability is generated according to the capabilities corresponding to the plurality of subtasks of the at least one subtask; 13. The method of any one of claims 2-12, wherein, receiving the fifth capability from the second device. The method further comprises:

14. The method of any one of claims 2-12, wherein, sending a sixth message to a third device, the sixth message comprising an identifier of the second capability corresponding to the first subtask, the sixth message being used to request authorization information of the second capability corresponding to the first subtask; receiving the authorization information of the second capability corresponding to the first subtask from the third device. The method further comprises:

15. The method of claim 14, wherein, sending a seventh message to the shared capability library, the seventh message comprising the second capability corresponding to any subtask of the at least one subtask, the seventh message being used to indicate that the second capability corresponding to the any subtask is added to the shared capability library. The method further comprises: determining a fifth capability according to the second capabilities corresponding to the plurality of subtasks of the at least one subtask, the fifth capability being used to implement the target task or to implement a part of the target task; 16. A method of communication, comprising: sending a seventh message to the shared capability library, the seventh message comprising the fifth capability, the seventh message being used to indicate that the fifth capability is added to the shared capability library. The method further comprises: sending an eighth message to the second device, the eighth message comprising the second capabilities corresponding to the plurality of subtasks of the at least one subtask, the eighth message being used to indicate that one capability is generated according to the capabilities corresponding to the plurality of subtasks of the at least one subtask; receiving the fifth capability from the second device. The method further comprises: sending a sixth message to a third device, the sixth message comprising an identifier of the second capability corresponding to the first subtask, the sixth message being used to request authorization information of the second capability corresponding to the first subtask; receiving the authorization information of the second capability corresponding to the first subtask from the third device. The method further comprises: sending a seventh message to the shared capability library, the seventh message comprising the second capability corresponding to any subtask of the at least one subtask, the seventh message being used to indicate that the second capability corresponding to the any subtask is added to the shared capability library. The method further comprises: determining a fifth capability according to the second capabilities corresponding to the plurality of subtasks of the at least one subtask, the fifth capability being used to implement the target task or to implement a part of the target task; sending a seventh message to the shared capability library, the seventh message comprising the fifth capability, the seventh message being used to indicate that the fifth capability is added to the shared capability library. The method further comprises: sending an eighth message to the second device, the eighth message comprising the second capabilities corresponding to the plurality of subtasks of the at least one subtask, the eighth message being used to indicate that one capability is generated according to the capabilities corresponding to the plurality of subtasks of the at least one subtask; receiving the fifth capability from the second device.

17. The method of claim 16, wherein, The receiving the task from the first device comprises: receiving a target task from the first device; The sending the first capability corresponding to the task to the first device comprises: sending a first capability corresponding to the target task to the first device.

18. The method of claim 16, wherein, The receiving the task from the first device comprises: receiving at least one subtask corresponding to a target task from the first device; The sending the first capability corresponding to the task to the first device comprises: sending second capabilities corresponding to the at least one subtask respectively to the first device, wherein the first capability comprises the second capabilities corresponding to the at least one subtask respectively.

19. The method of claim 17, wherein, The at least one subtask comprises a first subtask; The sending the second capability corresponding to the at least one subtask to the first device comprises: receiving a fifth message from the first device, wherein the fifth message comprises an identification of the second capability corresponding to the first subtask, and the fifth message is used to request the second capability corresponding to the first subtask; sending the second capability corresponding to the first subtask to the first device.

20. The method of claim 19, wherein, The method further comprises: deducting an execution cost of the second capability corresponding to the first subtask from an account corresponding to the first device.

21. The method of any one of claims 18-20, wherein, The method further comprises: receiving a seventh message from the first device, wherein the seventh message comprises a second capability corresponding to any one of the at least one subtask, and the seventh message is used to indicate that the second capability corresponding to the any one of the at least one subtask is added to the shared capability library; adding the second capability corresponding to the any one of the at least one subtask to the shared capability library.

22. The method of claim 21, wherein, The method further comprises: issuing a reward corresponding to the second capability corresponding to the any one of the at least one subtask to an account corresponding to the first device.

23. The method of any one of claims 18-20, wherein, The method further comprises: receiving a seventh message from the first device, wherein the seventh message comprises a fifth capability, and the fifth capability is generated according to second capabilities corresponding to a plurality of subtasks of the at least one subtask, and the seventh message is used to indicate that the fifth capability is added to the shared capability library; adding the fifth capability to the shared capability library.

24. The method of claim 23, wherein, The method further comprises: issuing a reward corresponding to the fifth capability to an account corresponding to the first device.

25. The method of any one of claims 16-24, wherein, The method further comprises: performing capability optimization on the capabilities in the shared capability library.

26. A communications device, characterized by comprises a processor and a memory; The processor is configured to execute computer programs or instructions stored in the memory, so that the communication device implements the method according to any one of claims 1-15, or implements the method according to any one of claims 16-25.

27. A computer-readable storage medium, characterized in that, The computer program or instruction is stored in the computer readable storage medium, and when the computer program or instruction is executed on the computer, the computer readable storage medium implements the method according to any one of claims 1-15, or implements the method according to any one of claims 16-25.

28. A chip, characterized by The chip comprises a processor coupled with a memory, for executing computer programs or instructions stored in the memory, so that the chip implements the method of any one of claims 1-15, or implements the method of any one of claims 16-25.

29. A communications device, characterized by The communication device comprises a module for executing the method of any one of claims 1-15, or the communication device comprises a module for executing the method of any one of claims 16-25.

30. A computer program product, characterised in that, When the computer reads and executes the computer program product, the computer implements the method of any one of claims 1-15, or implements the method of any one of claims 16-25.

Images