Method for generating multi-agent collaborative system through computing power of intelligent computing center

Abstract

The present invention provides a method for generating a multi-agent collaboration system through the computing power of an intelligent computing center, which is applied to a first multi-agent collaboration system, wherein the first multi-agent collaboration system includes a first team, and the first team includes a first agent and a second agent; the method includes: step S1: the first agent obtains a team building task given by a user in a first dialogue interface of the first multi-agent collaboration system, and the team building task is used to request the generation of a framework code for implementing a second multi-agent collaboration system, and the second agent collaboration system includes a second team; step S2: the first agent communicates with the user on the first dialogue interface to determine target demand information; step S3: the first agent instructs the second agent to generate the framework code of the second multi-agent collaboration system according to the target demand information; step S4: the second agent generates and stores the framework code of the second multi-agent collaboration system according to the target demand information.

Description

Technical Field

[0001] Embodiments of the present invention relate to the fields of computing power infrastructure and artificial intelligence technology, and in particular, to a method for generating a multi-agent collaboration system through the computing power of an intelligent computing center. Background Art

[0002] With the development of artificial intelligence technology and computing power technology, the concept of intelligent computing center has emerged. "Intelligent computing center" refers to the use of large-scale heterogeneous computing power resources, including general computing power and intelligent computing power, mainly for artificial intelligence applications (such as artificial intelligence deep learning model development, model training and model reasoning and other scenarios) to provide the required computing power, data and algorithms. Intelligent computing center covers facilities, hardware, software, and can provide full-stack capabilities from bottom-level computing power to top-level application enablement.

[0003] An agent is an agent that can perceive the environment and take actions to achieve specific goals. It can be software, hardware or a system with autonomy, adaptability and interaction. Agents perceive changes in the environment (such as through sensors or data input), make judgments and decisions based on their own learned knowledge and algorithms, and then perform actions to affect the environment or achieve predetermined goals. Agents are widely used in the field of artificial intelligence and are commonly seen in automated systems, robots, virtual assistants and game characters. The core of agents is the ability to learn autonomously and continuously evolve to better complete tasks and adapt to complex environments.

[0004] In related technologies, based on the powerful computing power of the intelligent computing center, more complex designated tasks can be completed through the collaboration of multiple intelligent agents. However, the development of a multi-agent collaborative system including multiple intelligent agents is a relatively complex process with high development costs, and the development process also fails to fully utilize the computing power of the intelligent computing center. Summary of the invention

[0005] An embodiment of the present invention provides a method for generating a multi-agent collaborative system through the computing power of an intelligent computing center, which is used to solve the problems in related technologies that the development process of the multi-agent collaborative system is complex, the development cost is high, and the development process fails to fully utilize the computing power of the intelligent computing center.

[0006] In order to solve the above-mentioned technical problems, the present invention is achieved as follows:

[0007] In a first aspect, an embodiment of the present invention provides a method for generating a multi-agent collaboration system by using computing power of an intelligent computing center, which is applied to a first multi-agent collaboration system, wherein the first multi-agent collaboration system includes a first team, and the first team includes a first agent and a second agent; the method includes:

[0008] Step S1: the first agent obtains a team building task given by a user in a first dialogue interface of the first multi-agent collaboration system, wherein the team building task is used to request generation of a framework code for implementing a second multi-agent collaboration system, wherein the second multi-agent collaboration system includes a second team for performing a specified task;

[0009] Step S2: the first agent communicates with the user through the first dialogue interface to determine target demand information, wherein the target demand information includes: information about the second multi-agent collaboration system, information about team members of the second team, collaboration relationships among the team members, and tool information of the team members, where each team member is an agent;

[0010] Step S3: the first agent instructs the second agent to generate a framework code for implementing the second multi-agent collaboration system according to the target requirement information;

[0011] Step S4: The second agent generates and stores a framework code for implementing the second multi-agent collaboration system based on the target demand information.

[0012] Optionally, step S4 further includes:

[0013] Step S41: the second agent obtains the storage address of the framework code, generates a download link according to the storage address, and displays the download link to the user through the first dialogue interface.

[0014] Optionally, the plurality of agents of the first team further include a third agent, and the method further includes:

[0015] Step S5: The third agent creates a remote development environment for editing the framework code, and displays access information of the remote development environment on the first dialogue interface.

[0016] Optionally, step S4 further includes:

[0017] Step S42: the second agent asks the user through the first dialogue interface whether to start the remote development environment;

[0018] Step S43: upon obtaining the information that the user needs to start the remote development environment, the second agent instructs the third agent to create a remote development environment.

[0019] Optionally, the information of the second multi-agent collaboration system includes: AppToken of the second multi-agent collaboration system;

[0020] and / or,

[0021] The information of the team members of the second team includes at least one of the following: the name of the team member, the brief introduction of the team member, the BotToken of the team member, and the number of the team members;

[0022] and / or,

[0023] The collaboration relationship of the team members includes: information of a collaboration relationship graph of the team members, wherein the collaboration relationship graph is a directed acyclic graph;

[0024] and / or,

[0025] The tool information of the team member includes: the name of the tool, the description of the tool, and the function declaration of the tool.

[0026] Optionally, the designated tasks include at least one of the following: large-scale language model training, computing power management of the intelligent computing center, financial reimbursement, patent mining, and travel planning.

[0027] In a second aspect, an embodiment of the present invention provides a multi-agent collaboration system, including a first team, the first team including a plurality of agents, the plurality of agents including a first agent and a second agent;

[0028] The first agent is used to obtain a team building task given by a user in a first dialogue interface of the multi-agent collaboration system, wherein the team building task is used to request generation of a framework code for implementing a second multi-agent collaboration system, wherein the second multi-agent collaboration system includes a second team for performing a specified task; communicate with the user through the first dialogue interface to determine target demand information, wherein the target demand information includes: information about the second multi-agent collaboration system, information about team members of the second team, collaborative relationships among the team members, and tool information of the team members, wherein each team member is an agent; and instruct the second agent to generate a framework code for implementing the second multi-agent collaboration system according to the target demand information;

[0029] The second agent is used to generate and store the framework code for implementing the second multi-agent collaboration system according to the target demand information.

[0030] In a third aspect, an embodiment of the present invention provides a server comprising: a processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the method for generating a multi-agent collaborative system through the computing power of an intelligent computing center as described in the first aspect above.

[0031] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for generating a multi-agent collaborative system through the computing power of an intelligent computing center as described in the first aspect above.

[0032] In a fifth aspect, an embodiment of the present invention provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the method for generating a multi-agent collaborative system through the computing power of an intelligent computing center as described in the first aspect above.

[0033] In the embodiment of the present invention, the first multi-agent collaboration system running in the intelligent computing center can collaboratively complete the task of generating a framework code for implementing the second multi-agent collaboration system in a dialogue interface through multiple agents in the first team. The user only needs to improve or modify the framework code to quickly complete the development of the second multi-agent collaboration system, which greatly reduces the development difficulty. In the process of generating the framework code for implementing the second multi-agent collaboration system, the powerful computing power of the intelligent computing center can be fully utilized to improve the processing speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0035] Figure 1 A flowchart of a method for generating a multi-agent collaboration system through the computing power of an intelligent computing center according to an embodiment of the present invention;

[0036] Figure 2-Figure 7 A schematic diagram of a dialogue interface for communicating needs between a first agent and a user according to an embodiment of the present invention;

[0037] Figure 8 A schematic diagram of generating a framework code for a second agent in an embodiment of the present invention;

[0038] Fig. 9 A schematic diagram of generating a remote development environment for a third agent in an embodiment of the present invention;

[0039] Fig.10 A directed acyclic graph corresponding to multiple agents of a multi-agent collaboration system according to an embodiment of the present invention;

[0040] Fig.11 is a schematic diagram of the structure of a server in an embodiment of the present invention;

[0041] Fig.12 Shown is the directory structure of the skeleton code for a project called devopsagent. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] First, the technical terms involved in the present invention are briefly explained below.

[0044] The "computing power" mentioned in the present invention is the ability of computer equipment or computing / data center to process information. It is the ability of computer hardware and software to work together to execute certain computing requirements. It is the computing power to achieve target result output by processing information data. It is a new type of productivity that integrates information computing power, network carrying capacity, and data storage capacity. It mainly provides services to the society through computing power infrastructure.

[0045] The "computing power" (Computational Power, CP) described in the present invention is the ability of a data center server to process data and output results. It is a comprehensive indicator to measure the computing power of a data center, including general computing power, super computing power and intelligent computing power. The commonly used unit of measurement is the number of floating point operations performed per second (FLOPS: Floating Point Operations Per Second, 1EFLOPS = 10^18FLOPS). The larger the value, the stronger the comprehensive computing power. According to calculations, 1EFLOPS is approximately the computing power output of 5 Tianhe-2A or 500,000 mainstream server CPUs or 2 million mainstream notebooks. The calculation formula is: CP = CP 通用 +CP 智能 +CP 超级 .

[0046] The "Network Power" (NP) described in the present invention is a manifestation of the data transmission capability of computing facilities, including comprehensive capabilities such as network architecture, network bandwidth, transmission latency, intelligent management and scheduling, etc. The carrying capacity involves network transmission within and between data centers, and is a comprehensive indicator for measuring network transmission scheduling capabilities. In the embodiment of the present invention, the carrying capacity uses the memory bandwidth.

[0047] The "Storage Power" (SP) described in the present invention is the comprehensive ability of a data center in terms of data storage capacity, performance, safety and reliability, and green and low-carbon. It is a comprehensive indicator for measuring the data storage capacity of a data center, including external storage devices such as storage arrays and server built-in storage devices. The commonly used unit of measurement for storage capacity is exabyte (EB, 1EB = 2^60bytes), and the commonly used unit of measurement for performance is the number of reads and writes per second per unit capacity (IOPS / TB, Input / Output Operations Per Second / TB). The disaster recovery ratio is an important manifestation of safety and reliability.

[0048] The "computing power infrastructure" described in the present invention is a new type of information infrastructure that integrates information computing power, network carrying capacity, and data storage capacity. It can realize centralized computing, storage, transmission and application of information, and presents characteristics such as diversity and ubiquity, intelligence and agility, security and reliability, and green and low-carbon.

[0049] The “computing power” mentioned in the present invention includes general computing power, intelligent computing power and super computing power.

[0050] The “general computing power” mentioned in the present invention refers to the computing power provided by servers based on CPU (Central Processing Unit) chips, which is used to support basic general computing such as cloud computing and edge computing.

[0051] The "intelligent computing power" described in the present invention is a computing platform for large-scale deployment of special chips such as GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), ASIC (Application Specific Integrated Circuit) for various innovative applications of artificial intelligence, such as natural language processing and machine vision.

[0052] The "super computing power" mentioned in the present invention is mainly the computing power provided by high-performance computing clusters such as supercomputers. It utilizes the centralized computing resources of multiple computer systems working in parallel and uses a dedicated operating system to handle extremely complex or data-intensive problems. It is mainly used for calculations in cutting-edge scientific fields, such as planetary simulation, drug molecule design, and gene analysis.

[0053] The "intelligent computing center" described in the present invention refers to a facility that provides the required computing power, data and algorithms for artificial intelligence applications (such as artificial intelligence deep learning model development, model training and model reasoning scenarios) by using large-scale heterogeneous computing power resources, including general computing power (CPU: Central Processing Unit) and intelligent computing power (GPU: Graphics Processing Unit, FPGA: Field Programmable Gate Array, ASIC: Application Specific Integrated Circuit, etc.). The intelligent computing center covers facilities, hardware, and software, and can provide full-stack capabilities from bottom-level computing power to top-level application enablement.

[0054] The "computing resources" mentioned in the present invention refer to the technologies and facilities with information calculation, transmission, storage and application capabilities required for the development of the digital society, including but not limited to computing resources such as CPU (Central Processing Unit), GPU (Graphics Processing Unit), network resources such as switches and routers, storage resources such as storage arrays and distributed storage, security resources such as firewalls and intrusion detection systems, as well as supporting and guarantee resources such as wind, fire, water and electricity.

[0055] The agent described in the embodiments of the present invention refers to an agent that can perceive the environment and take actions to achieve specific goals. It can be software, hardware or a system with autonomy, adaptability and interaction capabilities. The agent perceives changes in the environment (such as through sensors or data input), makes judgments and decisions based on the knowledge and algorithms it has learned, and then performs actions to affect the environment or achieve predetermined goals. Agents are widely used in the field of artificial intelligence and are commonly found in automated systems, robots, virtual assistants, and game characters. The core of agents is the ability to learn autonomously and continuously evolve to better complete tasks and adapt to complex environments.

[0056] The natural language processing (NLP) described in the embodiments of the present invention is a core technology that enables intelligent agents to understand, generate and interact with human natural language. Through NLP, intelligent agents can parse the semantics, grammar and context of human language, so as to perform tasks such as text understanding, language translation, dialogue system, sentiment analysis, etc. This enables intelligent agents to communicate with humans in natural language, understand the user's intentions, and provide corresponding responses and services.

[0057] The “large language model (also referred to as a large language model)” described in the embodiments of the present invention refers to a large language model (LLM), which is a language model with a large parameter scale that is designed to understand and generate human language. It is trained with a large amount of text data and can perform a wide range of tasks including text summarization, translation, sentiment analysis, etc.

[0058] The directed acyclic graph described in the embodiment of the present invention means that in graph theory, if a directed graph cannot start from a vertex and return to the vertex through several edges, then the graph is a directed acyclic graph (DAG graph).

[0059] To solve the problems in related technologies such as the complicated development process of multi-agent collaborative systems, high development costs, and the failure to fully utilize the computing power of the intelligent computing center during the development process, please refer to Figure 1 The embodiment of the present invention provides a method for generating a multi-agent collaboration system through the computing power of an intelligent computing center, which is applied to a first multi-agent collaboration system. The first multi-agent collaboration system includes a first dialogue interface and a first team. The first team includes multiple agents, each of which has specific capabilities. The multiple agents can collaborate in the first dialogue interface in a dialogue communication manner to complete a team building task of generating a framework code for implementing a second multi-agent collaboration system. The first dialogue interface is used to display the communication process between the user and the agent, the communication process between the agents, and part or all of the decision-making process of the agents during the execution of the task. The first team includes a first agent and a second agent. The method includes:

[0060] Step S1: the first agent obtains a team building task given by a user in a first dialogue interface of the first multi-agent collaboration system, wherein the team building task is used to request generation of a framework code for implementing a second multi-agent collaboration system, wherein the second multi-agent collaboration system includes a second team for performing a specified task;

[0061] Optionally, the designated task includes at least one of the following: large-scale language model training (also referred to as fine-tuning a large-scale speech model), computing power management of an intelligent computing center, financial reimbursement, patent mining, travel planning, etc.

[0062] Step S2: the first agent communicates with the user through the first dialogue interface to determine target demand information, wherein the target demand information includes: information about the second multi-agent collaboration system, information about team members of the second team, collaboration relationships among the team members, and tool information of the team members, where each team member is an agent;

[0063] Please refer to Figure 2-7 , Figure 2-7 Schematic diagram of a dialogue interface for communicating needs between a first agent and a user according to an embodiment of the present invention, Figure 2-7 It can be seen that the first agent (A) obtains the team building task given by the user (Zhang) in the dialogue interface (I want to form a team for fine-tuning a large model), and communicates with the user through the dialogue interface for multiple times to determine the target demand information.

[0064] Step S3: the first agent instructs the second agent to generate a framework code for implementing the second multi-agent collaboration system according to the target requirement information;

[0065] In the embodiment of the present invention, optionally, the first agent and the second agent can be different agents, that is, the first agent is responsible for communication needs, and the second agent is responsible for generating framework code. Of course, in some embodiments, they can also be the same agent, that is, the same agent completes communication needs and generates framework code.

[0066] Please refer to Figure 8 , Figure 8 A schematic diagram of generating a framework code for the second agent in an embodiment of the present invention. Figure 8 As can be seen, the first agent (A) is responsible for communication requirements, and the second agent (B) is responsible for generating framework code.

[0067] Step S4: The second agent generates and stores a framework code for implementing the second multi-agent collaboration system based on the target demand information.

[0068] Optionally, the framework code may include at least one of the following:

[0069] The first configuration file stores target demand information collected during the communication between the first agent and the user, such as the names of the team members of the second team, the profiles of the team members, the BotTokens of the team members, the tool information of the team members, etc.

[0070] The second configuration file is a startup configuration file, which is used in a development tool (such as VSCode) to define how to start an application (i.e., the application corresponding to the generated second multi-agent collaboration system). The development tool is used to develop the application corresponding to the second multi-agent collaboration system.

[0071] The first file contains functions or classes related to the team members (agents) of the second team.

[0072] The second file, also known as the main entry file, may contain the main logic of the application. After launching the file, the newly created agent team (i.e., the second team) may be used on a platform (e.g., the Slack platform).

[0073] The third file may include information such as an introduction to the second multi-agent collaboration system, instructions for use, and installation steps.

[0074] The fourth configuration file is used to store configuration information of the application, such as environment variables.

[0075] The fourth file is a dependency file for programming tools (such as Python), which lists all required programming tool packages and their versions.

[0076] For example, in one embodiment, please refer to Fig.12 , Fig.12 The following is a detailed explanation of each part:

[0077] 1.tmp / directory

[0078] This is a temporary directory, usually used to store temporary files or configuration files.

[0079] 1.1dev_config.yaml (corresponding to the first configuration file mentioned above): This is a configuration file, which stores the target demand information collected during the communication between the first agent and the user, such as the names of the team members of the second team, the profiles of the team members, the BotToken of the team members, the tool information of the team members, etc.

[0080] 1.2launch.json (corresponding to the second configuration file mentioned above): This is a startup configuration file, which is usually used in development tools (such as VSCode) to define how to start the application (that is, the application corresponding to the generated second multi-agent collaboration system).

[0081] 2. devopsagent / directory

[0082] This is the main code directory of the project.

[0083] 2.1experts / : This subdirectory contains all the expert function modules required for the newly created intelligent agent team (i.e. the second team).

[0084] 2.1.1init__.py: This is a Python package initialization file, which makes the experts directory a Python package.

[0085] 2.1.2 devops_engineer.py (corresponding to the first file above): This file contains the functions or classes related to the team members (agents) of the second team. The user only needs to open the tools to further implement the agent.

[0086] 2.2init__.py: This is the initialization file of the devops agent directory, making the entire directory a Python package.

[0087] 2.3app.py (corresponding to the second file mentioned above): This is the main entry file of the application, which can contain the main logic of the application. After starting this file, you can use the newly created intelligent team (ie, the second team) on the platform (Slack platform).

[0088] 3. Root directory files

[0089] 3.1README.md (corresponding to the third file above): This is a Markdown file containing project introduction, usage instructions, installation steps and other information.

[0090] 3.2Appconfig (corresponding to the fourth configuration file mentioned above): This is a configuration file used to store application configuration information, such as environment variables.

[0091] 3.3requirements.txt (corresponding to the fourth file above): This is a dependency file of a Python project, which lists all Python packages and their versions required by the project.

[0092] In the embodiment of the present invention, the first multi-agent collaboration system running in the intelligent computing center can collaboratively complete the task of generating a framework code for implementing the second multi-agent collaboration system in a dialogue interface through multiple agents in the first team. The user only needs to improve or modify the framework code to quickly complete the development of the second multi-agent collaboration system, which greatly reduces the development difficulty. In the process of generating the framework code for implementing the second multi-agent collaboration system, the powerful computing power of the intelligent computing center can be fully utilized to improve the processing speed.

[0093] At the same time, because specific demand scenarios require professional skills in this field, the multi-agent collaborative system cannot generate such skills. These skills may involve very professional knowledge. For example, financial accounting needs to be operated according to the company's specific financial platform or OA system. These may have special usage interfaces, etc. Therefore, professional personnel are needed to implement it. By using the multi-agent collaborative system to generate the framework code of another multi-agent collaborative system, and then the user can modify and improve the framework code by himself, the multi-agent collaborative system finally developed can be more adaptable to specific demand scenarios.

[0094] In the embodiment of the present invention, optionally, the information of the second multi-agent collaboration system includes: AppToken of the second multi-agent collaboration system;

[0095] In an embodiment of the present invention, optionally, the information of the team members of the second team includes at least one of the following: the name of the team member, the profile of the team member, the BotToken of the team member, and the number of team members; in an embodiment of the present invention, a team member is an intelligent agent.

[0096] In the embodiment of the present invention, optionally, the collaboration relationship of the team members includes: information of a collaboration relationship graph of the team members, the collaboration relationship graph being a directed acyclic graph;

[0097] In the embodiment of the present invention, optionally, the tool information of the team member includes: the name of the tool, the description of the tool, and the function declaration of the tool.

[0098] In some embodiments, the tool information of the team member only includes the function declaration of the tool, but does not include the specific implementation code of the tool. Therefore, the user also needs to improve the specific implementation code of the tool.

[0099] For example, one tool in the generated skeleton code contains the following code:

[0100]

[0101] When the tool is first generated, there is a pass in it. The user needs to implement the pass part with code, that is, what interface (search engine) to call to query the patent to be searched, and then topk similar patents.

[0102] That is, after the user completes the code of the tool, it may be:

[0103]

[0104] In some embodiments, the user's modification and improvement of the framework code may also include: modifying the information of the second multi-agent collaboration system in the framework code, such as modifying at least one of the following: the name of the team member, the profile of the team member, the number of team members, etc.

[0105] In the embodiment of the present invention, optionally, the step S4 further includes:

[0106] Step S41: the second agent obtains the storage address of the framework code, generates a download link according to the storage address, and displays the download link to the user through the first dialogue interface.

[0107] Please refer to Figure 8 , Figure 8 In the example, the second agent (B) obtains the storage address of the framework code (i.e. Figure 8 The code address in ), generates a download link according to the storage address (ie Figure 8 zip in the source file, and display the download link to the user through the first dialogue interface. Thus, the user can download the framework code to the local computer according to the download link, and open the framework code in a local editor (such as VS Code) to improve and modify the framework code.

[0108] In the embodiment of the present invention, optionally, the multiple agents of the first team further include a third agent, and the method further includes:

[0109] Step S5: The third agent creates a remote development environment for editing the framework code, and displays access information of the remote development environment on the first dialogue interface.

[0110] In the embodiment of the present invention, the access information may be an access address.

[0111] The remote development environment includes the framework code and configuration information of the framework code.

[0112] Please refer to Fig. 9 , Fig. 9 A schematic diagram of generating a remote development environment for a third agent in an embodiment of the present invention. Fig. 9 It can be seen that the third agent (C) creates a remote development environment for editing the framework code and displays the access information of the remote development environment on the first dialogue interface. Fig. 9 The seed code in is the framework code mentioned above.

[0113] According to the access information of the remote development environment, the user can access the remote development environment created on the remote server, and improve and modify the framework code on the remote server, thereby eliminating the need to configure the development environment locally.

[0114] In the embodiment of the present invention, optionally, the step S4 further includes:

[0115] Step S42: the second agent asks the user through the first dialogue interface whether to start the remote development environment;

[0116] Step S43: upon obtaining the information that the user needs to start the remote development environment, the second agent instructs the third agent to create a remote development environment.

[0117] In the embodiment of the present invention, optionally, the second agent and the third agent can be different agents, that is, the second agent is responsible for generating the framework code, and the second agent is responsible for generating the remote development environment. Of course, in some embodiments, they can also be the same agent, that is, the same agent completes the generation of the framework code and the generation of the remote development environment.

[0118] That is to say, in an embodiment of the present invention, two ways of obtaining the framework code can be provided. One way is that the user provides a download address and downloads the framework code to the local computer for modification and improvement. The other way is that the user connects to a remote server and directly modifies and improves the framework code on the server.

[0119] In the embodiment of the present invention, each agent in the first multi-agent collaboration system and the second multi-agent collaboration system is constructed based on a large language model, and each agent has specific capabilities. Each agent can communicate with other agents and users (i.e., humans, such as Figure 1 In the dialogue interface of the multi-agent collaborative system, the communication process between the user and the agent, the communication process between the agents, and part or all of the decision-making process of the agent can be displayed during the task execution.

[0120] In an embodiment of the present invention, the method may also include: the first agent obtains the capabilities of other agents in the first multi-agent collaborative system, and selects one or more agents from other second agents to form the first team based on the capabilities of the other agents.

[0121] In an embodiment of the present invention, the method may also include: the first intelligent agent obtains the capabilities of other intelligent agents in the first team, divides the team building task into multiple sub-tasks according to the capabilities of other intelligent agents in the first team, and assigns the multiple sub-tasks to other intelligent agents in the first team respectively.

[0122] In the embodiment of the present invention, optionally, the capabilities of the agent include decision-making capabilities and the ability to use tools. The decision-making capabilities are the decision-making capabilities possessed by the large language model corresponding to the agent. The tools may include the tools of the agent itself, and may also include external tools.

[0123] In an embodiment of the present invention, optionally, the first multi-agent collaborative collaboration and the multiple agents in the second multi-agent collaborative system are organized through a directed acyclic graph, wherein each node in the directed acyclic graph represents one of the agents, and each edge in the directed acyclic graph represents the communication between two of the agents located at both ends of the edge.

[0124] Please refer to Fig.10 , Fig.10 A directed acyclic graph corresponding to multiple agents of a multi-agent collaboration system according to an embodiment of the present invention, wherein the directed acyclic graph represents an organization with an overall goal, includes five nodes (A1, A2, A3, A4 and A5), each node represents an agent with a specific capability, and each edge in the directed acyclic graph represents the communication between two agents at both ends of the edge, wherein communication refers to sending information, such as Figure 4 As shown, A1 can communicate with A2, A3, A4, and A5, A2 can communicate with A1 and A4, and so on.

[0125] In the embodiment of the present invention, optionally, in the process of executing the specified task, each of the agents can follow the task instructions given by the agent at the previous level or the user, or give feedback suggestions to the agent at the previous level or the user. Fig.10 , Figure 4 Among them, A1 is the parent agent of A2, A3, A4 and A5, and A2 is the parent agent of A4.

[0126] In the embodiment of the present invention, optionally, agents at the same level can simultaneously execute the assigned tasks, thereby improving the execution efficiency of the tasks. Fig.10 , A2 and A3 are agents of the same level. A1 can assign tasks to A2 and A3, and A2 and A3 can perform the assigned tasks at the same time.

[0127] In the embodiment of the present invention, optionally, each of the intelligent agents can be used to perceive the user's dialogue instructions in real time during the execution of the specified task, and make decisions and take corresponding actions based on the user's dialogue instructions. That is to say, in the embodiment of the present invention, in the process of multiple intelligent agents collaborating to complete the task, humans can join in at any time to give suggestions on the execution of the specified task, or terminate the operation of the specified task, etc., which is highly flexible.

[0128] An embodiment of the present invention further provides a multi-agent collaboration system, comprising a first team, wherein the first team comprises a plurality of agents, wherein the plurality of agents comprises a first agent and a second agent;

[0129] The first agent is used to obtain a team building task given by a user in a first dialogue interface of the multi-agent collaboration system, wherein the team building task is used to request generation of a framework code for implementing a second multi-agent collaboration system, wherein the second multi-agent collaboration system includes a second team for performing a specified task; communicate with the user through the first dialogue interface to determine target demand information, wherein the target demand information includes: information about the second multi-agent collaboration system, information about team members of the second team, collaborative relationships among the team members, and tool information of the team members, wherein each team member is an agent; and instruct the second agent to generate a framework code for implementing the second multi-agent collaboration system according to the target demand information;

[0130] The second agent is used to generate and store the framework code for implementing the second multi-agent collaboration system according to the target demand information.

[0131] Optionally, the second agent is further used to obtain the storage address of the framework code, generate a download link according to the storage address, and display the download link to the user through the first dialogue interface.

[0132] Optionally, the multiple agents of the first team also include a third agent, and the third agent is used to create a remote development environment for editing the framework code and display access information of the remote development environment on the first dialogue interface.

[0133] Optionally, the second agent is further used to inquire the user through the first dialogue interface whether the user needs to start the remote development environment; if the user needs to start the remote development environment, instruct the third agent to create a remote development environment.

[0134] Optionally, the information of the second multi-agent collaboration system includes: AppToken of the second multi-agent collaboration system;

[0135] and / or,

[0136] The information of the team members of the second team includes at least one of the following: the name of the team member, the brief introduction of the team member, the BotToken of the team member, and the number of the team members;

[0137] and / or,

[0138] The collaboration relationship of the team members includes: information of a collaboration relationship graph of the team members, wherein the collaboration relationship graph is a directed acyclic graph;

[0139] and / or,

[0140] The tool information of the team member includes: the name of the tool, the description of the tool, and the function declaration of the tool.

[0141] Optionally, the designated tasks include at least one of the following: large-scale language model training, computing power management of the intelligent computing center, financial reimbursement, patent mining, and travel planning.

[0142] Please refer to Fig.11 The embodiment of the present invention further provides a server 110, including a processor 111, a memory 112, and a computer program stored in the memory 112 and executable on the processor 111. When the computer program is executed by the processor 111, each process of the above-mentioned method embodiment for generating a multi-agent collaboration system through the computing power of an intelligent computing center is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0143] The embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, each process of the above-mentioned method embodiment for generating a multi-agent collaboration system through the computing power of an intelligent computing center can be implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0144] The present application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the above Figure 1 The various processes of the method implementation example of generating a multi-agent collaboration system through the computing power of an intelligent computing center are shown, and the same technical effect can be achieved. To avoid repetition, they will not be described here.

[0145] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0146] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0147] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A method for generating a multi-agent collaborative system through the computing power of an intelligent computing center, characterized in that: Applied to a first multi-agent collaboration system, the first multi-agent collaboration system includes a first team, the first team includes a first agent and a second agent; the method includes: Step S1: the first agent obtains a team building task given by a user in a first dialogue interface of the first multi-agent collaboration system, wherein the team building task is used to request generation of a framework code for implementing a second multi-agent collaboration system, wherein the second multi-agent collaboration system includes a second team for performing a specified task; Step S2: the first agent communicates with the user through the first dialogue interface to determine target demand information, wherein the target demand information includes: information about the second multi-agent collaborative system, information about team members of the second team, collaborative relationships among the team members, and tool information of the team members, where each team member is an agent; Step S3: the first agent instructs the second agent to generate a framework code for implementing the second multi-agent collaboration system according to the target requirement information; Step S4: The second agent generates and stores a framework code for implementing the second multi-agent collaboration system based on the target demand information.

2. The method according to claim 1, characterized in that: The step S4 further comprises: Step S41: the second agent obtains the storage address of the framework code, generates a download link according to the storage address, and displays the download link to the user through the first dialogue interface.

3. The method according to claim 1, characterized in that The plurality of agents of the first team further include a third agent, and the method further includes: Step S5: The third agent creates a remote development environment for editing the framework code, and displays access information of the remote development environment on the first dialogue interface.

4. The method according to claim 3, characterized in that The step S4 further comprises: Step S42: the second agent asks the user through the first dialogue interface whether to start the remote development environment; Step S43: upon obtaining the information that the user needs to start the remote development environment, the second agent instructs the third agent to create a remote development environment.

5. The method according to claim 1, characterized in that: The information of the second multi-agent collaboration system includes: AppToken of the second multi-agent collaboration system; and / or, The information of the team members of the second team includes at least one of the following: the name of the team member, the brief introduction of the team member, the BotToken of the team member, and the number of the team members; and / or, The collaboration relationship of the team members includes: information of a collaboration relationship graph of the team members, wherein the collaboration relationship graph is a directed acyclic graph; and / or, The tool information of the team member includes: the name of the tool, the description of the tool, and the function declaration of the tool.

6. The method according to claim 1, characterized in that The designated tasks include at least one of the following: large-scale language model training, computing power management of intelligent computing centers, financial reimbursement, patent mining, and travel planning.

7. A multi-agent collaboration system, characterized in that: comprising a first team, the first team comprising a plurality of agents, the plurality of agents comprising a first agent and a second agent; The first agent is used to obtain a team building task given by a user in a first dialogue interface of the multi-agent collaboration system, wherein the team building task is used to request generation of a framework code for implementing a second multi-agent collaboration system, wherein the second multi-agent collaboration system includes a second team for performing a specified task; Communicate with the user through the first dialogue interface to determine target demand information, wherein the target demand information includes: information about the second multi-agent collaboration system, information about team members of the second team, collaboration relationships among the team members, and tool information of the team members, wherein each team member is an agent; instruct the second agent to generate a framework code for implementing the second multi-agent collaboration system according to the target demand information; The second agent is used to generate and store the framework code for implementing the second multi-agent collaboration system according to the target demand information.

8. A server, characterized in that: include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of a method for generating a multi-agent collaboration system through the computing power of an intelligent computing center as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for generating a multi-agent collaboration system through the computing power of an intelligent computing center as described in any one of claims 1 to 6.

10. A computer program product, characterized in that It includes computer instructions, which, when executed by a processor, implement the steps of the method for generating a multi-agent collaboration system through the computing power of an intelligent computing center as described in any one of claims 1 to 6.

Images