Global AI Agent inventory, big language model entrepreneurship must refer to 60 AI agents

• Global AI Agent inventory, 60 AI agents to help you understand AIGC innovation and entrepreneurship
• Global AI Agent inventory, big language model entrepreneurship must refer to 60 AI agents
• AI agents have become the first choice for AIGC entrepreneurship, and you need to understand this 60 world-renowned AI Agents
• AI agents into large models landing mainstream trend, what AI agents are worth referring to?
• To learn more about AI agents, these 60 AI agents are worth your in-depth study
• Global technology companies are exploring AI Agent, and AI agents may become a new standard for AIGC entrepreneurship

The full text is about 11,000 words and the reading time is 15 minutes

Text/Wang Jiwei

In April, shortly after Baidu's official release of Wen Xin Yiyan, many people were still lamenting how happy the pictures generated by Wen Xin Yiyan wen, and more people were crazy for ChatGPT, Midjourney various training, Meta founder and CEO Zuckerberg was thinking about how to introduce AI Agents to billions of people around the world in a "useful and meaningful way".

In May, when OpenAI completed its new $300 million round of funding, founder Sam Altman privately told some developers that he wanted to make ChatGPT a personal work assistant, and people familiar with the matter revealed that OpenAI has been focusing on how to use chatbots to create autonomous AI agents, and related functions are likely to be deployed in ChatGPT assistants.

At an all-hands meeting in June, Zuckerberg announced a series of technologies in various stages of development, one of which would bring AI Agents with different personalities and abilities to help or entertain users.

In China, although AutoGPT was synchronized with foreign countries as early as April, due to the lack of understanding of the AI Agent behind it by most people, the initial response was not too enthusiastic.

It was not until Lilian Weng, head of OpenAI applied artificial intelligence research, that the blog post about AI Agent exploded in the AI circle in early July, and the media circle, academic research community, and investment field really began to discuss AI Agent.

As a result, China has really opened a boom in exploring and researching AI Agents, and some manufacturers have begun to reconstruct product architecture and business models in AI Agent mode.

As the principles, patterns, and construction methods of AI agents become more and more inventive, many entrepreneurs trapped by technology, models, ecology and even policies are bright.

AI Agent not only allows everyone to see the direction of the landing of the large language model (LLM), so that more entrepreneurs can further ignite the hope of LLM entrepreneurship, but also allows the majority of enterprises to see the future trend of efficient application of LLM.

For AI Agent entrepreneurship, Andrej Karpathy, co-founder of OpenAI, believes that ordinary people, entrepreneurs and geeks have an advantage over OpenAI in building Agents, and everyone is in a state of equal competition.

On the side of large companies, in the face of the possibility of large technology companies and startups to seize this opportunity of Agent, Bill Gates also said that he would be disappointed if Microsoft did not intervene.

With the strong promotion of technology giants, the rapid embrace of entrepreneurs and the active introduction of large enterprises, AI Agent has become completely popular. And unlike the previous lack of LLM landing, this time AI Agent is no longer on paper, and many companies have launched Agent projects and related products.

According to industry insiders, at least 100+ projects are working to commercialize AI agents, and nearly 100,000 developers are building autonomous agents. Among these AI Agents, there are both foreign agent projects mainly based on GPT and open source agent frameworks, and domestic agent products based on domestic large models (self-developed field large models) + open source architecture.

Having said all that, which companies have launched Agent products? What is the current form of AI Agent products? In this article, Wang Ji Channel inventories sixty AI agents around the world to better understand AI agents.

PS: Due to the many agent items inventoried in this article, the word count has reached 1W+, it is recommended that you collect it first and then read it.

Start with AI Agent

Although LLM has enough intelligence, in order for it to give accurate answers, it needs to enter a precise enough prompt. A person who masters prompts and an ordinary person who asks questions using the same big model will get a very different answer: the former can use multiple techniques to get the desired result, while the latter can only look to LLM to sigh.

If you want to use LLM well, you must first learn to use prompt, and this demand has spawned a large training market. The prompt prompt project not only increases the difficulty of using LLM, but also reduces the user experience. LLM, which was supposed to show the advantages of natural language, became less user-friendly because of the complicated prompt.

In this way, the prompt project has become a mountain between ordinary people and large models.

How to better solve this problem? The answer is AI Agent (known as AI agent in China).

An AI agent is an intelligent entity that perceives the environment, makes decisions, and performs actions. Different from traditional AI, AI Agent has the ability to gradually complete a given goal by thinking independently and invoking tools.

After the arrival of LLM, AI agents were defined as LLM-driven agents to automate the processing of common problems.

We know that LLM is mainly good at processing and generating text. They can answer questions, write articles, generate creative content, help with programming, and more. But LLM is also a passive tool that only produces output when you give it input.

AI agents offer a wider range of capabilities, especially in terms of interacting with the environment, proactive decision-making, and performing various tasks. It can be said that AI Agent is the key to truly unlocking the potential of LLM, which can provide powerful action capabilities for the LLM core.

The main difference between AI agents and large models is that the interaction between large models and humans is based on prompts. Whether the user's prompt is clear and unambiguous will affect the effect of the large model's answer, and there is no accurate and effective prompt, even the most capable ChatGPT.

The work of the AI agent only needs to be given a goal, it can think and act independently for the goal, it will break down each step of the planning step in detail according to the given task, rely on feedback from the outside world and independent thinking, and create a prompt for itself to achieve the goal.

For example, if you ask ChatGPT to buy a cup of coffee, the feedback given by ChatGPT is generally similar to "can't buy coffee, it's just a text AI assistant" and the like.

But you have to tell the ChatGPT-based AI Agent tool to buy a cup of coffee, it will first disassemble how to buy a cup of coffee for you and plan to use an APP to place an order and pay for several steps, and then follow these steps to call the APP to select takeaway, and then call the payment program to place an order payment, the process does not require humans to specify each step of the operation.

While both AI tools and agents are software programs designed to automate tasks, specific key characteristics distinguish AI agents into more complex AI software.

An AI tool can be considered an AI agent when it has the following characteristics:

• Autonomy: AI virtual agents are able to perform tasks independently without human intervention or input.
• Perception: The agent function senses and interprets the environment in which they are located through various sensors, such as cameras or microphones.
• Reactivity: AI agents can evaluate the environment and respond accordingly to achieve their goals.
• Reasoning and decision-making: AI agents are intelligent tools that analyze data and make decisions to achieve goals. They use reasoning techniques and algorithms to process information and take appropriate action.
• Learning: They can learn and improve their performance through machines, deep and reinforcement learning elements, and techniques.
• Communication: AI agents can communicate with other agents or humans using different methods, such as understanding and responding to natural language, recognizing speech, and exchanging messages via text.
• Goal-oriented: They aim to achieve specific goals that can be predefined or learned through interaction with the environment.

In terms of categories, AI agents can currently be divided into autonomous agents and generative agents.

Autonomous agents such as Auto-GPT are able to automate tasks and achieve desired outcomes based on the needs people make through natural language. In this cooperative model, the autonomous agent is primarily at the service of humans and is more like an efficient tool.

Generate agents, such as the Westworld town co-created by researchers at Stanford and Google or the humanoid robots in Westworld, that live in the same environment, have their own memories and goals, and interact not only with humans but also with other robots.

Regarding AI agents, the 86-page LLM-based Agents review paper recently launched by the Natural Language Processing Team of Fudan University (FudanNLP) comprehensively sorts out the current situation of intelligent agents based on large-scale language models, including: the background, composition, application scenarios of LLM-based Agents, and the agent society that has attracted much attention.

(PS: Interested friends, you can send a message agent in the background to get the paper, as usual, Wang Jiwei's channel has prepared a machine copy for everyone.) ）

Having said all this, many friends may still not have intuitive feelings about AI agents. Don't worry, below we will deepen everyone's understanding through a comparative case.

AI agents permeate all fields

AiAgent.app is a web application that allows users to create custom AI agents to perform specific tasks and achieve goals.

The following Wang Jiwei channel will compare the experience of using AI agents and directly using LLM, and look at the advantages of AI agents.

For example, if you want to know the news and trends of the AI industry in the past month, enter in Claude: a summary of the latest news and trends in the AI industry in the past month.

The results obtained are shown below:

As you can see, Claude just lists a few summaries of AI-related news.

In AiAgent.app entering this paragraph, it first breaks down your requirements into ten tasks, then interacts with the user through prompts to complete each task and outputs the results for each task. Obviously, the content obtained in the AiAgent.app about the recent AI industry is more comprehensive than the content obtained directly with other LLMs.

Can I get this directly using the big model? Theoretically, you can do this by typing more prompts, but you need to enter them at least ten times, you don't guarantee the accuracy of the incoming prompts, and sometimes you don't even know what information you want to get.

In AiAgent.app, you only need to enter a sentence, it analyzes your possible needs and lists relatively comprehensive content goals, guiding you to complete what you want, and increasing efficiency several times.

Compared with the two, in terms of the richness and efficiency of content acquisition, AI Agent is obviously superior. This kind of information content agent has great value for media practitioners, industry analysts and other professions, and can greatly reduce the time to obtain research materials.

There are already some such agents for more precise user groups and use cases, such as GPT Researcher launched by Columbia University, a ChatGPT-based investigator-oriented agent that can create various research reports for users to promote research.

This case is only the acquisition of content, in fact, there are now agents for multiple application scenarios, which are enough to mobilize more software applications and even hardware devices to complete various tasks.

For example, some people have already used AutoGPT to order, book, taxi and shop; The 25 AI agents in Stanford's Westworld town walk, date, chat, drink coffee, and share the day's news every day; Google Deepmind launched a robotic agent that uses robotic arms to automate various tasks; Amazon also launched Amazon Bedrock Agents to automatically break down enterprise AI application development tasks. IBM Watson Health already helps doctors diagnose, treat and monitor patients in many hospitals.

Although Ai Agent has not been popular for a long time, it has been supported by many companies in many fields as soon as it appeared. The multi-model capability of the large language model and the greater computing power today allow the agent proposed many years ago to quickly gain value and land in more fields with a super penetration rate.

With the emergence of open source AI agents such as MetaGPT, more technology vendors and entrepreneurial teams introduce Agents, and more organizations recognize and accept Agents, which will inevitably quickly become the main mode of LLM landing in various fields, helping thousands of industries better apply LLM.

A large inventory of 60 AI agents around the world

The AiAgent.app mentioned in the above case is one of the representative products of AI Agent that has been in the limelight in recent months. Many agents at home and abroad, including this AI agent, can be seen in the project inventory list below.

In order to better understand the AI Agents that have been launched, Wang Jiwei's channel divides these AI Agents into media reports, domestic launches, industry-based, overseas and GitHub projects, and will gradually reward the project library in the future, and classify these Agents into different categories.

The AI Agents listed in this article include both AI Agents frameworks and tools, as well as AGENT products based on some open source frameworks, and most of the projects and products are autonomous agents.

Because some manufacturers are relatively low-key and have not been publicized, the AI Agent inventoried in this article is not complete, so it is also called the AI AGENT incomplete list. Welcome more manufacturers and entrepreneurs to contact Wang Jiwei Channel after seeing this article, and everyone will make a contribution to the prosperity and development of the AI AGENT ecosystem.

First, the AI agent reported by the media

1、Auto-GPT

Auto GPT is a free and open source project on Github that combines GPT-4 and GPT-3.5 technologies to create complete projects through APIs.

Unlike ChatGPT, users don't need to constantly ask the AI questions to get answers, in AutoGPT they only need to give them an AI name, description, and five goals, and AutoGPT can complete the project itself. It can read and write files, browse the web, review the results of its own prompts, and combine it with said hint history.

Auto-GPT is one of the first examples of GPT-4 operating completely autonomously, pushing the boundaries of what AI can do.

2、AgentGPT

AgentGPT allows you to configure and deploy autonomous AI agents. Just give your custom AI a name and let it start with any goal imaginable, and it can try to achieve it by thinking about the task to be done, performing the task, and learning from the results.

3、Baby AGI

It's an AI-driven task management system. The system uses OpenAI and the Pinecone API to create, prioritize, and execute tasks. Create tasks by analyzing the results of previous tasks and predefined targets, and use OpenAI's natural language processing (NLP) and Chroma to store and retrieve task results in context.

The appeal of Baby AGI lies in its ability to autonomously solve tasks and maintain predefined goals based on the results of previous tasks, as well as effectively prioritizing tasks.

4、Jarvis （HuggingGPT）

A unique collaboration system developed by Microsoft that uses multiple AI models to accomplish a given task, with ChatGPT acting as a task controller. The project, dubbed JARVIS on GitHub, is now available for trial on Huggingface (hence the name HuggingGPT), an agent that works very well with text, images, audio, and even video.

It works similarly to OpenAI's multimodal capabilities that showcase GPT 4 through text and images, but JARVIS further integrates a variety of open source LLMs for images, video, audio, and more, as well as connecting to the internet and accessing files. For example, you can enter a URL from a website and ask relevant questions.

5、Aiagent.app

Ai Agent is a web application that allows users to create custom AI agents to perform specific tasks and achieve goals. AI agents work by breaking down targets into smaller tasks and completing them one by one. Benefits include the ability to run multiple AI agents simultaneously and democratize access to cutting-edge technology.

AI Agent also boasts features such as inline code blocks with syntax highlighting, and seamless collaboration with third-party platforms. The tool is free to use, and it provides a simplified way to build AI agents without the need for more technical knowledge.

6、Camel AGI

Camel AGI is a generative AI tool that enables users to solve a given task by role-playing autonomous AI agents, of course users need to enable Javascript to use this tool. Camel AGI allows users to complete tasks using AI agents and offers the option to log in with Google or star the tool on Github.

7. "Westworld" simulation Westworld town

Researchers from Stanford University and Google created an interactive sandbox environment with 25 generative AI agents that can mimic human behavior. They walk in parks, drink coffee in cafes, and share news with colleagues, showing surprisingly good social behavior.

For example, starting with a user-specified concept that an agent wants to host a Valentine's Day party, the agent automatically spreads the party invitation over the next two days, meets new people, asks each other on a date party, and coordinates to appear together at the party at the right time.

8、GPT-Engineer

GPT-Engineer is an open-source AI tool that allows users to specify what they want to build and then have a clarifying conversation with the AI to generate the desired code base. The tool is designed to provide a simple and flexible user experience, allowing users to adapt and extend their functionality to their needs.

The tool includes functions such as specifying the identity of the AI agent, storing the communication history with GPT4, and rerunning the message log. Contributions to the project are welcome, and interested individuals can refer to the roadmap, projects, and issues available on the GitHub repository. GPT-Engineer aims to be an open platform for developers to explore and build their code generation toolboxes.

9、MetaGPT

MetaGPT is an open-source multi-agent framework that uses a single line of input to generate APIs, user stories, data structures, competitive analysis, and more. The framework can act as a product manager, software engineer, and architect. The framework can act as an entire software company, orchestrating SOPs with just one line of code.

MetaGPT is integrated with human SOP process design. As a result, LLM-based agents generate high-quality, diverse, structured documents and designs. MetaGPT's design makes it easy to design solutions for complex tasks and provides problem-solving capabilities that are almost comparable to human intelligence.

10、Amazon Bedrock Agents

Amazon Bedrock Agents, released by Amazon, allows developers to quickly create fully managed agents. By making API calls to enterprise systems, Amazon Bedrock agents accelerate the release of generative AI applications that can manage and execute activities.

Amazon Bedrock Agents simplifies the rapid engineering and orchestration of user-requested tasks. Once set up, these agents can build prompts on their own and securely enhance them with company-specific data to provide natural language responses to users. These advanced agents have the ability to infer the necessary actions to automatically process user requests.

11、nvidia Voyager

Voyager, co-launched by NVIDIA, California Institute of Technology, etc., uses GPT-4 to guide the learning Minecraft agent through the pixel world, it should be noted that Voyager relies on code generation, not reinforcement learning.

Voyager is the first lifelong learning agent to play Minecraft. Unlike other Minecraft agents that use classical reinforcement learning techniques, Voyager uses GPT-4 to continuously improve itself, doing so by writing, improving, and transferring code stored in external skill bases.

This spawns mini-programs that help you navigate, open doors, mine resources, craft pickaxes, or fight zombies. GPT-4 unlocks a new paradigm in which "training" is the execution of code and "training model" is Voyager's iteratively assembled skill code base.

12、RoboAgent

The joint research team of Meta and CMU spent two years successfully developing the RoboAgent universal robot agent. RoboAgent has achieved 12 different complex skills through training on just 7,500 tracks, including tasks such as baking, picking up items, serving tea, cleaning the kitchen, etc., and can be generalized in 100 unknown scenarios.

No matter how much distraction it encounters, RoboAgent stays on track. The goal of this research is to build an efficient robotic learning paradigm that addresses the challenge of diversity of datasets and scenarios. The researchers propose a multi-task action chunked Transformer (MT-ACT) architecture to process multimodal multi-task robot datasets through semantic enhancement and efficient policy representation.

13、Inflection AI Pi

Inflection AI's personal AI Agent product Pi, the core brain is the company's research and development of the Inflection-1 large model, the performance is comparable to GPT-3.5. Pi is different from the popular general-purpose chatbots in that it can only engage in friendly conversations, offer concise advice, or even just listen.

Its main characteristics are compassionate, humble and curious, humorous and innovative, with good emotional intelligence, and can provide unlimited knowledge and companionship according to the unique interests and needs of users. Since the beginning of the development of the Pi, Inflection has determined that the Pi will serve as a personal intelligence, not just a tool to assist people in their work.

14、HyperWrite

Hyperwrite is an AI writing agent tool that helps creative writers of any level write faster and with more confidence. It includes features like auto-writing and typing ahead to generate original passages and come up with ideas to overcome writers' obstacles.

The tool comes as a free Chrome extension that can be used on any website without interrupting the workflow. It is used and trusted by professionals, students, and creators around the world to increase their productivity.

15、GPT Researcher

GPT Researcher is an AI-based autonomous agent for comprehensive online research on a variety of tasks. Inspired by AutoGPT and "plan and solve" prompts, the tool aims to improve the speed and deterministic problems identified in current language models, "providing more stable performance and higher speed through parallel agent work, rather than synchronous operation."

According to the team, GPT researchers facilitate research by generating relevant research questions, aggregating data from more than 20 web sources, and leveraging GPT3.5-turbo-16 and GPT-4 to create comprehensive research reports.

The AI Agent that has been launched in China

After continuous exploration and attempts, domestic AI agent-related products have also begun to emerge, and the following five products are introduced.

1. Alibaba Cloud ModelScopeGPT

Alibaba Cloud's Mota community launched the first large-scale model invocation tool in China (ModelScopeGPT), through this tool, users can send instructions to call other artificial intelligence models in the Mota community with one click, so as to achieve large and small models to work together to complete complex tasks.

ModelScopeGPT is based on ModelScope-Agent, an AI Agent development framework for open source Large Language Model (LLM). This is a general-purpose, customizable Agent framework for real-world applications, based on open source Large Language Models (LLMs) as the core, including modules for memory control, tool usage, and more.

Open source LLM is mainly responsible for task planning, scheduling, and reply generation; Memory control module, mainly including knowledge retrieval and prompt management; Tool usage modules, including tool libraries as well as tool retrieval and tool customizability.

2. Truly intelligent TARS-RPA-Agent

The TARS-RPA-Agent launched by Real Intelligence in the field of hyperautomation is a new RPA model product based on the "TARS+ISSUT (Intelligent Screen Semantic Understanding)" dual-mode engine, with "brain" and "eyes and hands and feet", which is a new RPA model product that can independently disassemble tasks, perceive the current environment, execute and feedback, and remember historical experience.

TARS-RPA-Agent adopts a technical framework based on the TARS large model and ISSUT intelligent screen semantic understanding. The technical framework is divided into two layers: the bottom layer is the TARS series of large models including the general basic model and the basic model of various vertical industries, and the intelligent screen semantic understanding technology; The upper layer is a hyper-automated product that has been comprehensively upgraded and transformed based on these two key technologies.

The core LLM of TARS-RPA-Agent is a self-developed vertical "TARS" large model based on the base of the general large model, and the TARS large model has excellent mainstream capabilities such as text generation, language understanding, knowledge question answering, and logical reasoning.

3, OmBot ohm agent

At the 2023 World Artificial Intelligence Conference, Lianhui Technology released the OmBot Ohm agent, an autonomous agent based on large model capabilities, and launched the first batch of applications for typical scenarios.

As an automatic, autonomous agent, it runs in a loop in its simplest form, generating self-directed instructions and actions with each iteration. As a result, it does not rely on humans to guide commands and is highly scalable.

4. Ask XBot

The agent platform "Ask XBot" built by Lanma Technology is divided into two layers: the first layer is expert empowerment, experts define the workflow through dragging, pulling, dragging and dialogue interaction, and teach the machine, so as to assist front-line employees to build a methodology for more efficient work; The second layer is that employees use Agent, and front-line employees can communicate with the Agent through natural language and give instructions, so that the Agent can assist in data analysis and data retrieval.

The company plans to build Ask XBo into a platform that is both versatile and easy to use, managing these APIs and agents, allowing agents to wrap different APIs, and agents of different models can better collaborate on it, so that they can serve customers more efficiently and intelligently on the platform.

5、ChatDev

ChatDev, launched by a joint research team of Tsinghua University, Beijing University of Posts and Telecommunications, and Brown University, is a generative agent. It is a chat-based end-to-end software development framework that leverages Large Language Models (LLMs) to facilitate effective communication and collaboration between multiple roles ("GPT's "gpt3.5-turbo-16k" version of ChatGPT) in the software development process.

The main purpose of ChatDev is game development through chat. Users simply come up with ideas, and the entire process from design to testing is done by AI, and the whole process can be completed in just seven minutes.

AI Agent products for different fields

Before the advent of LLM, some companies were already studying the combination of traditional AI and Agent. Therefore, the implementation of AI Agenmt in various fields is much faster than everyone expected.

Below are representative agent applications in several industry fields.

• In the medical field, agents can help diagnose, treat, and monitor patients. IBM Watson Health is an AI agent that analyzes medical data to identify potential health problems and recommend treatment options.
• In the financial sector, agents can analyze financial data, detect fraud, and make investment recommendations. Charles Schwab uses an AI agent called Intelligent Portfolio to create and manage portfolios based on clients' investment goals.
• In retail business scenarios, agents can provide personalized recommendations, improve supply chain management, and enhance customer experience. Amazon's Alexa is an AI agent that can recommend products, place orders, and track shipments.
• In manufacturing, agents can optimize production processes, predict maintenance needs, and improve product quality. GE uses an AI agent called Predix to monitor machines in real time to predict and prevent equipment failures.
• In the transportation sector, autonomous AI agents can assist with route planning, traffic management, and vehicle safety. Tesla's Autopilot helps self-driving vehicles and helps drivers park, change lanes and drive safely.
• In the education industry, Agnet can provide a personalized learning experience, automate administrative tasks and analyze student performance. Pearson's AI agent, Aida, can provide students with feedback and suggest personalized learning paths.
• In agriculture, AI agents can optimize crop production, monitor soil quality, and predict weather patterns. John Deere is using an AI agent called See & Spray to detect and locate weeds without affecting crops.

Other AGENT products have been launched overseas

1、Cognosys

Cognosys is a web-based AI agent designed to revolutionize productivity and simplify complex tasks, enhancing your daily life using state-of-the-art AI technology.

2、Doanythingmachine

Easily manage your tasks with an "omnipotent" machine, and the user's personal AI agent will prioritize and complete your tasks for you

3、alphakit

Intuitive platform for creating and managing goal-driven autonomous AI agent teams, all from your mobile autoGPT AI agent teams. Just define your goals and Alphakit takes care of the rest.

4、GPTConsole

GPTConsole is a revolutionary command-line interface (CLI) designed to give developers the benefits of artificial intelligence. It goes beyond traditional terminal functionality to enable users to perform complex tasks using prompts.

5、Finished

Provide links to your knowledge base to turn your knowledge base into an AI chat in 2 minutes. Fini provides users with a tireless AI agent ready to answer customer questions instantly 24/7.

6、Spell

Spell is an autonomous AI agent based on GPT4, which can be applied to daily efficient work. Spell also has much-needed features to help you work smarter and learn to harness the power of generative AI to generate one or more innovative autonomous agents that will work to solve your problems.

7、Aomni

Aomni is an information retrieval AI agent that can find, extract and process any data on the Internet for you, enhancing your research efforts. Aomni can use a variety of tools to intelligently plan your queries to get the final result, including a full web browser that allows it to access any information on the internet without the need for an API.

Based on the current state-of-the-art AutoGPT architecture, Aomni's query planner intelligently schedules and updates each request to ensure the correctness and diversity of sources.

8、Fine-Tuner.ai

With Fine-Tuner.ai, users can build complex, tailor-made AI agents without technical skills or coding, just by entering your data and ideas. More than a dozen professional AI agents can create accurate Q&A, document search, process automation, etc. for users through uploaded instant data such as PDF, CV, PPT, URL, etc.

9、SuperAGI

An open-source autonomous AI framework that enables you to quickly and reliably develop and deploy useful autonomous agents for building, managing, and operating the infrastructure of autonomous agents.

10、Yellow.ai

Yellow.ai is the leading enterprise-grade conversational AI platform that powers enterprise dynamic AI agents and aims to deliver human-like interactions through its no-code/low-code platform to improve customer satisfaction and employee engagement.

11、Godmode

Enables users to run AutoGPT in a browser. Godmode allows users to deploy multiple AI agents simultaneously to complete tasks using AI, or they can use their own OpenAI API keys.

12、E42

E42 is a cognitive process automation platform that companies can use to create multifunctional cognitive agents to automate various processes across functions. Cognitive-driven, no-code platforms integrate seamlessly with users' existing technologies and processes to unlock the highest value across departments. Users can use E42 to build their own AI agents, such as AI analysts and AI recruiters across vertical industries.

13、Thankful

Thankful's AI agents are trained and tailored to work within your existing help desk, easily resolving a large number of customer queries via email, chat, SMS, and in-app channels. With the ability to understand, connect, solve, personalize, and inform, ThankfulAI agents deliver human-like service experiences at machine-like speed and innately scalable expertise.

14、Aktify

Clone your sales team with Aktify's virtual AI agent without increasing headcount. Aktify will handle an unlimited number of unresponsive leads at scale) and consistently bring conversation-ready customers to your sales team's door, and it's more than just an SMS chatbot.

15、TeamSmart AI

Increase your productivity with one-click access to TeamSmart AI. Aggregate content, generate code, draft tweets, and more directly in the browser. Click on an icon or keyboard shortcut to open ChatGPT instantly, giving you instant access to the library of quality tips without logging in.

16、BrainstormGPT

BrainstormGPT integrates multiple agents, LLM, and auto-search to simplify topic-to-session report conversion. Custom topics, user-defined roles, agent self-discussions, reports output in 20 minutes, approximately 300 searches, 10 hours of discussions, and 100,000 text analytics.

17、AgentRunner.Ai

AgentRunner.ai is an autonomous AI agent creation tool that harnesses the power of GPT-4 to create and train fully autonomous agents. Allow users to set goals for their agents and let them decide how to achieve those goals without any technical knowledge or programming skills.

The features offered by the tool include creating autonomous agents with unique personalities, running agents to perform tasks or learn new skills, deciding what agents can do, and integrating with OpenAI or Google Cloud accounts.

18、Stay

Gista helps businesses engage with website visitors and convert them into leads 24/7, and its main features include building AI transformation agents and AI sales agents. With Gista, businesses can easily convert website visitors into leads and build an email list.

19、Agent4

One of the main features of Agent4 is the ability to create AI-powered virtual agents that can answer questions, help book meetings, listen to voicemails, and provide summaries.

You can easily create custom interactions for agents, enabling them to answer questions and handle a variety of tasks with your brand's voice. You can also choose how agents respond to calls in real time and decide if and when they need to talk to someone.

20、Cometcore AI

Cometcore AI is an innovative platform that offers a versatile range of AI-driven tools to improve productivity and communication. With Cometcore, you can make, code, and automate cute agents.

21、personal-assistant

An AI agent designed to handle everything from booking flights to conducting in-depth research and everything in between.

AI Agent project on Github

1、OpenAGI

OpenAGI is an open-source AGI research platform specifically designed to deliver complex multi-step tasks with task-specific datasets, evaluation metrics, and various scalable models. OpenAGI formulates complex tasks as natural language queries as input to LLM. LLM then selects, synthesizes, and executes the model provided by OpenAGI to solve the task.

The project also proposes a Task Feedback Reinforcement Learning (RLTF) mechanism that uses task solving results as feedback to improve LLM's task solving ability. LLM is responsible for synthesizing various external models to solve complex tasks, while RLTF provides feedback to improve its task solving capabilities, providing a feedback loop for self-improving AI. The paradigm of LLM manipulating various expert models to solve complex tasks is a promising approach for AGI.

2、Agent-LLM

Agent-LLM is an AI automation platform designed to power efficient AI instruction management across multiple providers.

The agent is equipped with adaptive memory, a versatile solution that provides a powerful plug-in system that supports a variety of commands, including web browsing. With increasing support for numerous AI providers and models, Agent-LLM continues to evolve to enhance a variety of applications.

3、AutoGPT-Next-Web

The agent enables one-click deployment of a well-designed AutoGPT-Next-Web Web UI on Vercel and a free one-click deployment of your private AutoGPT-Next-Web web application. Based on AutoGPT-Next-Web, users can use Vercel to deploy for free and build a personal AutoGPT website within 1 minute.

4、MiniGPT-4

This agent can enhance visual language understanding using advanced large language models.

5、Mini-AGI

Mini-AGI is the smallest general-purpose autonomous agent based on GPT 3.5/4. It combines powerful prompts, a minimal set of tools and short-term memory (chain of thoughts), augmented by vector-stored data that will soon be added, to analyze stock prices, perform cybersecurity tests, create artwork, and order pizza.

6、Teenage-AGI

This intelligence project was inspired by several Auto-GPT-related projects (mainly BabyAGI) and the paper "Generative Agents: Interactive Simulations of Human Behavior", this Python project uses OpenAI and Pinecone to provide memory for AI agents and allow it to "think" before taking action (output text).

7、FastGPT

FastGPT is a knowledge base question answering system based on LLM large language model, which provides out-of-the-box data processing, model invocation and other capabilities. At the same time, workflow orchestration can be performed through Flow visualization, enabling complex Q&A scenarios

8、DemoGPT

With DemoGPT, you can quickly create a presentation using simple sentences.

9、LocalAGI

Local running AGI projects based on LLMDA, ChatGLM and other models.

10, AI-Town (games)

a16z open-source AI Town, an MIT-licensed, deployable starter kit for building and customizing your own version of AI Town. It's a virtual town where AI characters live, chat, and socialize.

11. GPTRPG (Games)

gptrpg This repository contains two things: a simple RPG-like environment for an AI agent that supports LLM; Simple AI agents that connect to OpenAI APIs to exist in that environment.

12, SFighterAI (games)

This project is an AI agent trained using deep reinforcement learning to defeat the final boss in the game Street Fighter II: Special Champion Edition. The AI agent makes decisions based only on the RGB pixel value of the game screen. In the saved state provided, the agent reaches a 100% win rate in the first round of the final level.

Written at the end: Wang Jiwei's channel continues to pay attention to the latest progress of AI agents at home and abroad, and this list of AI agents will continue to improve. Welcome companies, teams and entrepreneurs who have launched AI agents to join me on WeChat to discuss and exchange AIGC entrepreneurship and innovation.

End of full text

[Wang Jiwei channel, pay attention to AIGC and IoT, focus on digital transformation, business process automation and RPA, welcome to follow and communicate. 】