Source: Wang Jiwei

Image Source: Generated by Wujie AI

• A comprehensive inventory of global AI Agents, 60 AI intelligent entities to help you understand AIGC innovation and entrepreneurship
• A comprehensive inventory of global AI Agents, 60 AI intelligent entities that entrepreneurs in large language models must refer to
• AI intelligent entities become the first choice for AIGC entrepreneurship, you need to understand these 60 globally renowned AI Agents
• AI intelligent entities become the mainstream trend of large model landing, what AI Agents are worth referring to?
• If you want to delve into AI intelligent entities, these 60 AI Agents are worth your in-depth research
• Global technology companies are exploring AI Agents, AI intelligent entities may become the new standard for AIGC entrepreneurship

Shortly after Baidu officially released Wenxin Yiyuan in April, while many people were still marveling at the joyful images generated by Wenxin Yiyuan, and many more were going crazy for various training sessions for ChatGPT and Midjourney, Meta's founder and CEO Mark Zuckerberg was thinking about how to "introduce AI Agents to billions of people around the world in a useful and meaningful way."

In May, OpenAI completed a new round of $300 million in financing. Founder Sam Altman privately told some developers that he hopes to turn ChatGPT into a personal work assistant. Insiders revealed that OpenAI has been focusing on how to use chatbots to create autonomous AI Agents, and these features are very likely to be deployed in the ChatGPT assistant.

At a company-wide meeting in June, Zuckerberg announced a series of technologies at different stages of development, one of which will bring AI Agents with different personalities and abilities to help or entertain users.

In China, although AutoGPT had already gained popularity in sync with foreign countries in April, due to the lack of understanding of the AI Agent behind it, the initial response was not too enthusiastic.

It wasn't until early July when OpenAI's AI research leader Lilian Weng's blog post about AI Agents went viral in the AI community that the media, academic research, and investment fields truly began to fervently discuss AI Agents.

As a result, China has truly embarked on a craze of exploring and researching AI Agents, and some companies have begun to restructure their product architecture and business models using the AI Agent model.

As the principles, models, and construction methods of AI Agents become clearer, many entrepreneurs who have been constrained by technology, models, ecosystems, and even policies are excited.

AI Agents not only show the direction of large language model (LLM) landing, igniting further hope for LLM entrepreneurship among more entrepreneurs, but also show the future trend of efficient application of LLM to a wide range of enterprises.

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

On the other hand, large companies, facing the opportunity of both large tech companies and startups seizing the Agent, Bill Gates also expressed disappointment if Microsoft did not get involved.

With the strong promotion of tech giants, rapid embrace by entrepreneurs, and active introduction by large enterprises, AI Agents have become extremely popular. Unlike the previous situation where LLM lacked practical application, this time AI Agents are no longer just talk on paper, and many companies have already launched Agent projects and related products.

Industry insiders have revealed that at least 100+ projects are dedicated to commercializing AI intelligent entities, and nearly 100,000 developers are building autonomous Agents. Among these AI Agents, there are projects mainly based on GPT and open-source Agent frameworks from abroad, as well as Agent products based on domestic large models (self-developed large model) and open-source frameworks in China.

With all that said, which companies have launched Agent products? What are the current forms of AI Agent products? This article on the Wang Ji channel lists 60 AI Agents worldwide to help everyone better understand AI intelligent entities.

Starting with AI Agents

Although LLM has sufficient intelligence, in order to get precise answers from it, it is necessary to input a precise prompt. There will be a significant difference in the answers obtained by a person who masters the prompt and an ordinary person using the same large model to ask questions: the former can use various techniques to get the desired results, while the latter can only hope for the best from LLM.

To use LLM effectively, one must first learn to use prompts, which has given rise to a considerable training market. Prompt engineering, while increasing the difficulty of using LLM, also reduces the user experience. LLM, which should have showcased its advantage in natural language, has become less user-friendly due to complex prompts.

In this way, prompt engineering has become a huge barrier between ordinary people and large models.

How can this problem be better addressed? The answer lies in AI Agents (referred to as AI intelligent entities in China).

AI Agents are intelligent entities capable of perceiving the environment, making decisions, and taking actions. Unlike traditional AI, AI Agents have the ability to independently think, call tools, and gradually achieve given goals.

After the arrival of LLM, AI Agents were defined as Agent implementations driven by LLM to automate general problems.

We know that LLM is mainly good at processing and generating text. They can answer questions, write articles, generate creative content, and help with programming, among other things. However, LLM is still a passive tool that only produces output when you input something to it.

AI Agents provide a wider range of functions, especially in interacting with the environment, making active decisions, and executing various tasks. It can be said that AI Agents are the key to truly unleashing the potential of LLM, providing powerful action capabilities for the core of LLM.

The main difference between AI Agents and large models is that the interaction between large models and humans is based on prompts. The clarity and precision of the user's prompt will affect the effectiveness of the large model's response. Without a precise and effective prompt, even the most powerful ChatGPT cannot perform well.

The work of AI Agents only requires a given goal, and it can independently think and take action based on the goal. It will break down the detailed plan for each step of the given task, create prompts for itself to achieve the goal, and rely on feedback from the outside world and independent thinking.

For example, if you ask ChatGPT to buy a cup of coffee, the feedback from ChatGPT is usually something like "unable to buy coffee, it is just a text AI assistant."

However, if you instruct an AI Agent tool based on ChatGPT to buy a cup of coffee, it will first break down how to buy a cup of coffee for you and plan out several steps, such as using a certain app to place an order and make a payment, and then it will call the app to select delivery, and then call the payment program to place the order and make the payment, without the need for human intervention in each step.

Although both AI tools and Agents are software programs aimed at automating tasks, specific key features distinguish AI intelligent entities as more complex AI software.

When an AI tool has the following features, it can be considered an AI Agent:

• Autonomy: AI virtual intelligent entities can independently execute tasks without the need for human intervention or input.
• Perception: Intelligent entities perceive and interpret their environment through various sensors, such as cameras or microphones.
• Reactivity: AI intelligent entities can assess the environment and make corresponding responses to achieve their goals.
• Reasoning and decision-making: AI intelligent entities are intelligent tools that can analyze data and make decisions to achieve goals. They use reasoning techniques and algorithms to process information and take appropriate actions.
• Learning: They can learn and improve their performance through machine, deep, and reinforcement learning elements and techniques.
• Communication: AI intelligent entities can communicate with other intelligent entities or humans using different methods, such as understanding and responding to natural language, recognizing speech, and exchanging messages through text.
• Goal-oriented: They are designed to achieve specific goals, which can be predefined or learned through interaction with the environment.

In terms of categories, AI intelligent entities can currently be divided into Autonomous Agents and Generative Agents.

Autonomous Agents, such as Auto-GPT, can automatically execute tasks and achieve expected results based on people's natural language requests. In this cooperative mode, Autonomous Agents primarily serve humans and are more like efficient tools.

Generative Agents, such as the Westworld town created jointly by researchers from Stanford and Google, or the humanoid robots in "Westworld," live in the same environment, have their own memories and goals, interact not only with humans but also with other robots.

Regarding AI intelligent entities, a recent 86-page review paper on LLM-based Agents was released by Fudan University's Natural Language Processing Team (FudanNLP), comprehensively outlining the current status of intelligent agents based on large language models, including the background, composition, application scenarios, and the highly anticipated agent society of LLM-based Agents.

After all this, many friends may still not have a clear understanding of AI intelligent entities. Don't worry, in the following text, we will deepen everyone's understanding through a comparative case study.

Penetration of AI Intelligent Entities in Various Fields

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

The Wang Jiwei channel will use a comparative experience of using Ai intelligent entities and directly using LLM to see the advantages of AI intelligent entities.

For example, if you want to understand the latest news and trends in the AI industry over the past month, you can input the following into Claude: "Summary of the latest news and trends in the artificial intelligence industry in the past month."

The results obtained are as follows:

As you can see, Claude only listed a few summaries of AI-related news and information.

On the other hand, when you input this sentence into AiAgent.app, it will first break down your request into ten tasks, then interact with the user through prompts to complete each task, and provide results for each task. Obviously, the content about the recent AI industry obtained from AiAgent.app is more comprehensive than what you would get directly from using other LLM.

Can you obtain this content directly using a large model? Theoretically, it can be done by inputting more prompts, but it would require at least ten inputs, and there is no guarantee of prompt accuracy, and sometimes you may not even know what information you want to obtain.

On the other hand, with AiAgent.app, you only need to input one sentence, and it analyzes your potential needs and lists out relatively comprehensive content goals, guiding you to achieve what you want, greatly increasing efficiency.

In comparison, in terms of content richness and efficiency, it is clear that AI Agents have the upper hand. This type of information content Agent is of great value to media professionals, industry analysts, and other professions, as it can greatly reduce the time spent on research material acquisition.

There are already some Agents tailored to more specific user groups and application scenarios, such as GPT Researcher launched by Columbia University, which is an Agent for researchers based on ChatGPT and can create various research reports to promote research.

This case is just about content acquisition. In fact, there are now Agents for multiple application scenarios, enough to mobilize more software applications and even hardware devices to complete various tasks.

For example, some people have already used AutoGPT to order meals, book tickets, take taxis, and shop; the 25 AI Agents in the Westworld town at Stanford walk, date, chat, have coffee, and share the day's news every day; Google Deepmind has launched robotic agents that use mechanical arms to automatically perform various tasks; Amazon has also launched Amazon Bedrock Agents for automatically decomposing enterprise AI application development tasks; IBM Watson Health has helped doctors diagnose, treat, and monitor patients in many hospitals.

Although the time when AI Agents became popular is not very long, they have been embraced by many enterprises in various fields. The multi-modal capabilities of large language models, combined with even greater computing power today, have quickly highlighted the value of Agents proposed many years ago, and with their super high penetration rate, they are landing in more fields.

With the emergence of open-source AI Agents such as MetaGPT, more technology suppliers and entrepreneurial teams are introducing Agents, and more organizations are recognizing and accepting Agents. It will undoubtedly become the main mode of landing for LLM in various fields, helping thousands of industries to better apply LLM.

Comprehensive Inventory of 60 AI Agents Worldwide

The AiAgent.app mentioned in the above case is one of the representative products of AI Agents that has been popular in recent months. Several Agents from both domestic and international sources, including this AI intelligent entity, can be seen in the list of projects below.

To help everyone better understand the currently launched AI Agents, the Wang Jiwei channel will categorize these AI Agents into media reports, domestic launches, industry types, overseas others, and GitHub projects, and will gradually expand the project library to categorize these Agents into different categories.

The AI Agents listed in this article include AI Agent frameworks and tools, as well as AGENT products built on some open-source frameworks, and most projects and products are for autonomous intelligent entities.

Because some companies are relatively low-key and have not publicized their products, the AI Agents listed in this article are not exhaustive, so it is also called the incomplete list of AI Agents. More companies and entrepreneurs are welcome to contact the Wang Jiwei channel after seeing this article, and together we can contribute to the prosperity and development of the AI Agent ecosystem.

AI Agents Reported in the Media

1. Auto-GPT

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

Unlike ChatGPT, where users need to continuously ask questions to AI to get corresponding answers, in Auto GPT, users only need to provide an AI name, description, and five goals, and then Auto GPT can complete the project on its own. It can read and write files, browse the web, review its own prompt results, and combine them with the history of prompts.

Auto-GPT is one of the first examples of fully autonomous operation of GPT-4, breaking the limits of what artificial intelligence can do.

2. AgentGPT

AgentGPT allows you to configure and deploy autonomous AI intelligent entities. Simply name your custom AI and let it start any imaginable goal, and it will attempt to achieve the goal by thinking about the task to be completed, executing the task, and learning from the results.

3. Baby AGI

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

The appeal of Baby AGI is its ability to autonomously solve tasks and maintain predefined goals based on the results of previous tasks, as well as effectively determine task priorities.

4. Jarvis (HuggingGPT)

Developed by Microsoft, Jarvis is a unique collaborative system that uses multiple AI models to complete given tasks, with ChatGPT acting as the task controller. The project is called JARVIS on GitHub and is now available for trial on Huggingface (hence called HuggingGPT), and this Agent works very well with text, images, audio, and even video.

It operates similarly to OpenAI's demonstration of GPT-4's multimodal capabilities through text and images, but JARVIS further integrates various open-source LLMs for images, videos, audio, and can also connect to the internet and access files. For example, you can input a URL from a website and ask related questions.

5. Aiagent.app

Ai Agent is a web application that allows users to create custom AI intelligent entities to perform specific tasks and achieve goals. The working principle of AI Agents is to break down goals into smaller tasks and complete them one by one. The benefits include the ability to run multiple AI intelligent entities simultaneously and democratize access to cutting-edge technology.

AI Agent also features inline code blocks with syntax highlighting and seamless collaboration with third-party platforms. The tool is free to use and provides a simplified way to build AI intelligent entities without requiring extensive technical knowledge.

6. Camel AGI

Camel AGI is a generative AI tool that allows users to solve given tasks through role-playing with autonomous AI intelligent entities, and users need to enable JavaScript to use this tool. Camel AGI allows users to complete tasks with AI intelligent entities and provides options to log in using Google or star the tool on GitHub.

7. "Westworld" simulation

This project comes from researchers at Stanford University and Google, who have created an interactive sandbox environment containing 25 generative AI intelligent entities capable of simulating human behavior. They walk in the park, have coffee in the café, and share news with colleagues, demonstrating surprisingly good social behavior.

For example, starting from a concept specified by a user, that an intelligent entity wants to host a Valentine's Day party, the intelligent entity automatically spreads party invitations, meets new friends, mutually invites each other to the party, and coordinates to appear at the party at the right time over the next two days.

8. GPT-Engineer

GPT-Engineer is an open-source AI tool that allows users to specify the content they want to build and engage in clarifying conversations with AI to generate the required codebase. The tool aims to provide a simple and flexible user experience, allowing users to adjust and expand its functionality according to their needs.

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

9. MetaGPT

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

MetaGPT integrates with human SOP process design. Therefore, based on LLM, intelligent entities generate high-quality, diverse, and structured documents and designs. The design of MetaGPT 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's Amazon Bedrock Agents allow developers to quickly create fully managed intelligent entities. By executing API calls to enterprise systems, Amazon Bedrock intelligent entities accelerate the deployment of generative AI applications that can be managed and executed.

Amazon Bedrock Agents simplify rapid engineering and orchestration of user-requested tasks. Once set up, these intelligent entities can autonomously generate prompts and enhance prompts with company-specific data securely, providing natural language responses to users. These advanced intelligent entities have the ability to autonomously handle the necessary operations for user-requested tasks.

11. nvidia Voyager

Voyager, launched by NVIDIA, Caltech, and others, uses GPT-4 to guide learning Minecraft intelligent entities through the pixel world. It is important to note that Voyager relies on code generation rather than reinforcement learning.

Voyager is the first lifelong learning intelligent entity to play "Minecraft." Unlike other Minecraft intelligent entities using classic reinforcement learning techniques, Voyager continuously improves itself using GPT-4 by writing, improving, and transmitting code stored in an external skill library.

This produces small programs to help with navigation, opening doors, mining resources, making pickaxes, or fighting zombies. GPT-4 unlocks a new paradigm in which "training" is the execution of code, and "training the model" is the iterative assembly of skill code libraries by Voyager.

12. RoboAgent

Developed by a joint research team from Meta and CMU, RoboAgent is a general-purpose robot intelligent entity that successfully developed 12 different complex skills with just 7500 trajectories of training, including baking, picking up items, making tea, and cleaning the kitchen, and can generalize application in 100 unknown scenes.

Regardless of the level of interference, RoboAgent can persist in completing tasks. The goal of this research is to establish an efficient robot learning paradigm to address the challenges of dataset and scene diversity. The researchers proposed a multi-task action chunk Transformer (MT-ACT) architecture to handle multimodal multi-task robot datasets through semantic enhancement and efficient policy representation.

13. Inflection AI Pi

Inflection AI Pi

Inflection AI's personal AI Agent product Pi, with the core brain being the company's developed Inflection-1 large model, performs on par with GPT-3.5. Unlike popular general chatbots, Pi can engage in friendly conversations, provide concise advice, and even just listen.

Its main features include empathy, humble curiosity, humor, and innovation, with high emotional intelligence, providing unlimited knowledge and companionship based on the user's unique interests and needs. Since the development of Pi, Inflection has determined that Pi will serve as a Personal Intelligence, rather than just a tool to assist human work.

14. HyperWrite

Hyperwrite is an AI writing intelligent entity tool that helps creative writers of any level write faster and more confidently. It includes features such as automatic writing and predictive typing, generating original paragraphs and suggesting ideas to overcome writer's block.

The tool is available as a free Chrome extension and can be used on any website without interrupting the workflow. It is used and trusted by professionals, students, and creators worldwide to enhance their productivity.

15. GPT Researcher

GPT Researcher is an AI-based autonomous intelligent entity for comprehensive online research on various tasks. Inspired by AutoGPT and "plan and solve" prompts, the tool aims to improve the speed and determinism issues found in current language models by providing more stable performance and higher speed through parallel intelligent entity work rather than synchronous operations.

According to the team, GPT Researcher facilitates research by generating relevant research questions, summarizing data from over 20 web resources, and using GPT3.5-turbo-16 and GPT-4 to create comprehensive research reports.

Domestic AI Agents

After continuous exploration and experimentation, domestic AI intelligent entity-related products have begun to emerge. Below are introductions to five such products.

1. Alibaba Cloud ModelScopeGPT

Alibaba Cloud's Mota community launched the first large-scale model invocation tool in China, ModelScopeGPT, which allows users to call other artificial intelligence models in the Mota community with a single command, enabling models of all sizes to collaborate and complete complex tasks.

ModelScopeGPT is based on the ModelScope-Agent, a development framework for AI Agents based on open-source large language models (LLMs). It is a universal and customizable Agent framework for practical applications, with the core based on open-source large language models (LLMs), including modules for memory control, tool usage, and more.

2. Real Intelligence TARS-RPA-Agent

Real Intelligence's TARS-RPA-Agent, the first in the field of hyper-automation, is a new mode product based on the "TARS+ISSUT (Intelligent Screen Semantic Understanding)" dual-mode engine, with a "brain," "eyes, and hands" that can autonomously break down tasks, perceive the current environment, execute and provide feedback, and remember historical experiences.

TARS-RPA-Agent is based on the TARS large model and ISSUT intelligent screen semantic understanding technology. The technology framework consists of a bottom layer with TARS series large models and intelligent screen semantic understanding technology, and a top layer that completes comprehensive upgrades and transformations of hyper-automation products based on these two key technologies.

3. OmBot

At the 2023 World Artificial Intelligence Conference, Lianhui Technology released the OmBot autonomous intelligent entity based on large model capabilities, and launched the first batch of applications tailored to typical scenarios.

Lianhui's autonomous intelligent entity includes four core capabilities: cognition, memory, thinking, and action. As an automatic and autonomous intelligent entity, it runs in a loop in the simplest form, generating self-directed instructions and operations at each iteration. Therefore, it does not rely on human guidance and is highly scalable.

4. Lanma Technology Ask XBot

Lanma Technology has built the Agent platform "Ask XBot," which consists of two layers: the first layer empowers experts to define workflows through drag-and-drop and dialogue interaction, teaching the machine to assist frontline workers in building more efficient work methodologies. The second layer allows frontline workers to communicate with the Agent using natural language and issue commands to assist with tasks such as data analysis and document retrieval.

The company plans to develop Ask XBot into a platform that combines universality and ease of use, effectively managing these APIs and Agents, allowing different models of Agents to collaborate more efficiently and intelligently on the platform to better serve customers.

5. ChatDev

ChatDev, developed by a joint research team from Tsinghua University, Beijing University of Posts and Telecommunications, and Brown University, is a generative intelligent entity based on a chat-based end-to-end software development framework. It facilitates effective communication and collaboration among multiple roles (ChatGPT's "gpt3.5-turbo-16k" version) using large language models (LLMs) in the software development process.

The main purpose of ChatDev is to develop games through chat. Users only need to present ideas, and the entire process from design to testing is completed by AI in just seven minutes.

AI Agents in Different Fields

Before the emergence of LLM, some enterprises were already researching the combined application of traditional AI and Agents. Therefore, the landing of AI Agents in various fields has been much faster than expected.

Below are representative Agent applications in several industry sectors.

• In the medical field, Agents can help diagnose, treat, and monitor patients. IBM Watson Health is an AI intelligent entity that can analyze medical data to identify potential health issues and recommend treatment plans.
• In the financial sector, Agents can analyze financial data, detect fraudulent behavior, and provide investment advice. Charles Schwab uses an artificial intelligence intelligent entity called Intelligent Portfolio to create and manage investment 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 intelligent entity that can recommend products, place orders, and track shipments.
• In the manufacturing industry, Agents can optimize production processes, predict maintenance needs, and improve product quality. General Electric uses an AI intelligent entity 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 with autonomous driving and assists drivers with parking, lane changes, and safe driving.
• In the education industry, Agents can provide personalized learning experiences, automate administrative tasks, and analyze student performance. Pearson's AI intelligent entity Aida can provide feedback to students and suggest personalized learning paths.
• In the agricultural field, AI Agents can optimize crop production, monitor soil quality, and predict weather patterns. John Deere is using an AI intelligent entity called See&Spray to detect and locate weeds without affecting crops.

Other AGENT Products Launched Overseas

1. Cognosys

Cognosys is a web-based AI intelligent entity designed to revolutionize productivity and simplify complex tasks using advanced AI technology to enhance daily life.

2. Doanythingmachine

Easily manage your tasks with the "Doanythingmachine," where your personal AI intelligent entity will prioritize and complete your tasks for you.

3. alphakit

An intuitive platform for creating and managing goal-driven autonomous AI intelligent entity teams, all created and managed through a mobile autoGPT AI intelligent entity team. Just define your goals, and Alphakit takes care of the rest.

4. GPTConsole

GPTConsole is a revolutionary command-line interface (CLI) designed to provide the advantages of artificial intelligence to developers, allowing users to execute complex tasks using prompts.

5. Fini

Fini provides links to knowledge bases, converting your knowledge base into AI chat in 2 minutes. Fini provides users with a tireless AI intelligent entity ready to answer customer questions 24/7.

6. Spell

Spell is a GPT4-based autonomous AI intelligent entity for efficient daily work. Spell also has essential features to help you work smarter and learn to harness the powerful capabilities of generative AI to create innovative autonomous intelligent entities dedicated to solving your problems.

7. Aomni

Aomni is an information retrieval AI intelligent entity that can find, extract, and process any data on the internet to enhance your research work. Aomni can intelligently plan your queries using various tools to obtain the best results, including a full web browser that allows access to any information on the internet without the need for an API.

8. Fine-Tuner.ai

With Fine-Tuner.ai, users can build complex, customized AI intelligent entities without technical skills or coding, simply by inputting their data and ideas. Dozens of professional AI intelligent entities can create precise question-and-answer, document search, and process automation based on uploaded instant data such as PDFs, CVs, PPTs, URLs, and more.

9. SuperAGI

An open-source autonomous AI framework that allows you to quickly and reliably develop and deploy useful autonomous intelligent entities for building, managing, and running the infrastructure of autonomous intelligent entities.

10. Yellow.ai

Yellow.ai is a leading enterprise conversational AI platform that supports dynamic AI intelligent entities for enterprises, aiming to improve customer satisfaction and increase employee engagement through human-like interactions using its no-code/low-code platform.

11. Godmode

Allows users to run AutoGPT in the browser. Godmode allows users to deploy multiple AI intelligent entities to complete tasks using AI and also use their own OpenAI API key.

12. E42

E42 is a cognitive process automation platform that allows enterprises to create multifunctional cognitive intelligent entities for automating various processes across functions. The cognitive-driven no-code platform seamlessly integrates with users' existing technology and processes to unleash the highest value across departments. Users can use E42 to build their own AI intelligent entities, such as AI analysts and cross-vertical AI recruiters.

13. Thankful

Thankful's AI intelligent entity, trained and tailored, can easily handle a large volume of customer queries through existing help desks via email, chat, SMS, and in-app channels. With the ability to understand, connect, resolve, personalize, and notify, Thankful's AI intelligent entity provides a human-like service experience with machine-like speed and inherently scalable expertise.

14. Aktify

Using Aktify's virtual AI intelligent entity to clone your sales team without increasing staff numbers. Aktify efficiently handles an unlimited number of unresponsive potential customers and consistently brings ready-to-talk customers to your sales team's door, more than just a text chatbot.

15. TeamSmart AI

Improve your work efficiency with one-click access to TeamSmart AI. Summarize content, generate code, draft tweets directly in the browser. Click the icon or use keyboard shortcuts to immediately open ChatGPT without logging in for instant access to quality prompts.

16. BrainstormGPT

BrainstormGPT integrates multiple intelligent entities, LLM, and automatic search to simplify the transformation of topics into meeting reports. Custom topics, user-defined roles, autonomous intelligent entity discussions, and reports output in 20 minutes, equivalent to about 300 searches, 10 hours of discussion, and 100,000 text analyses.

17. AgentRunner.Ai

AgentRunner.ai is a tool for creating autonomous AI intelligent entities that uses the powerful capabilities of GPT-4 to create and train fully autonomous intelligent entities. It allows users to set goals for their intelligent entities and let them decide how to achieve those goals without any technical knowledge or programming skills.

18. Gista

Gista

Gista can help businesses interact with website visitors and convert them into 24/7 potential customers, with key features including building AI conversion agents and AI sales agents. With Gista, businesses can easily convert website visitors into potential customers and build email lists.

19. Agent4

One of Agent4's key features is the ability to create AI-driven virtual agents that can answer questions, assist with booking meetings, listen to voicemails, and provide summaries. You can easily create custom interactions for agents, allowing them to respond to calls in your brand's voice and handle various tasks. You can also choose how agents respond to calls in real-time and decide if and when a conversation with someone is needed.

20. Cometcore AI

Cometcore AI is an innovative platform that offers a range of multifunctional AI-driven tools to enhance productivity and communication. With Cometcore, you can create, code, and automate intelligent agents.

21. personal-assistant

An artificial intelligence agent designed to handle tasks ranging from booking flights to conducting in-depth research and everything in between.

AI Agent Projects on GitHub

1. OpenAGI

OpenAGI is an open-source AGI research platform designed to provide complex multi-step tasks with task-specific datasets, evaluation metrics, and various scalable models. OpenAGI represents complex tasks as natural language queries, serving as input for LLM. LLM then selects, synthesizes, and executes models provided by OpenAGI to solve tasks. The project also introduces the Task Feedback Reinforcement Learning (RLTF) mechanism, which uses task-solving results as feedback to improve LLM's task-solving capabilities.

2. Agent-LLM

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

3. AutoGPT-Next-Web

This agent can deploy a carefully designed AutoGPT-Next-Web Web UI on Vercel with a single click, allowing users to build their personal AutoGPT website within 1 minute using Vercel's free one-click deployment.

4. MiniGPT-4

This agent enhances visual language understanding using advanced large language models.

5. Mini-AGI

Mini-AGI is a minimal general autonomous intelligent entity based on GPT3.5/4. It combines powerful prompts, a minimal set of tools, and short-term memory (thought chains) with vector-stored data enhancement, allowing analysis of stock prices, execution of network security tests, creation of artwork, and ordering pizza.

6. Teenage-AGI

This intelligent project, inspired by several Auto-GPT-related projects (mainly BabyAGI) and the paper "Generative Agents: Interactive Simulation of Human Behavior," uses OpenAI and Pinecone to provide memory for AI intelligent entities and allows them to "think" before taking action (outputting text).

7. FastGPT

FastGPT is a knowledge base question-answering system based on LLM large language models, providing out-of-the-box data processing, model invocation, and the ability to orchestrate complex question-answering scenarios through visual Flow.

8. DemoGPT

With DemoGPT, you can quickly create demos using simple sentences.

9. LocalAGI

A local AGI project based on models such as LLMDA and ChatGLM.

10. ai-town (Gaming)

ai-town, an open-source AI town from the renowned investment firm a16z, is a MIT-licensed deployable starter kit for building and customizing your own version of an AI town. It is a virtual town where AI characters live, chat, and socialize.

11. gptrpg (Gaming)

This repository contains a simple RPG-like environment to support AI intelligent entities using LLM, connected to the OpenAI API to exist within the environment as simple AI intelligent entities.

12. SFighterAI (Gaming)

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

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。