Devin AI is the world's first fully autonomous AI programming software development programmer developed by Cognition AI. Its core strength lies in its ability to solve complex problems independently in a sandbox computing environment.Devin AI's performance has been rigorously tested in SWE-bench coding benchmarks, demonstrating its excellence in solving real-world problems.Devin is an AI programmer with a full-stack of skills, the ability to autonomously learn new technologies, build and deploy apps end to end, and autonomously find and fix code bugs, AI programmer who trains and fine-tunes AI models. It leverages the...

April 17, 2011 - OpenAI today released Codex CLI, a lightweight end-run coding intelligence that is now fully open source on GitHub. Codex CLI works directly on the user's computer and is designed to maximize the inference capabilities of models such as o3 and o4-mini, with support for additional API models such as GPT-4.1 coming soon. Users can gain multimodal inference capabilities from the command line, such as by passing screenshots or low-fidelity sketches to the model, combined with access to native code...

April 10 news, the State Administration of Radio and Television (SARFT) said in an article today, the State Administration of Radio and Television (SARFT) Academy of Radio and Television Science and Research actively carry out research on the application of generative artificial intelligence technology, integration of text, image, video, audio and other artificial intelligence models, research and development of broadband television and audio intelligences (AI Agent) development tools, has completed DeepSeek, Ali Tongyi Wanxiang, Tencent Mixed Meta, Step Leap The integration of multiple domestic open source big models such as DeepSeek, Ali Tongyi Wanxiang, Tencent Mixed Yuan, and Step Leap Star has been completed. 1AI learned from the official introduction that the tool realizes the whole process of program planning and creation of intelligence, with topic planning, script creation, script sub-scene, human...

First, let's look at what AutoGLM contemplation is. It means that you only need to issue a command, you can go to wash the dishes and walk the cat. I. How to use [AutoGLM contemplation] intelligent body 1. Open the homepage of Wisdom Spectrum Clear Speech: https://www.1ai.net/2443.html The normal way of using it is just like you use other big language models. However, we are here to experience the [AutoGLM Contemplation] Intelligent Body. 2. You can click Tutorial Guide first to see how to use the tutorial. https://zhipu-a...

April 2, 2011 - In a Twenty Minute VC podcast that aired on March 31st local time, Microsoft CTO Kevin Scott predicted that "in the next year, AI agents' biggest flaw -- their ability to remember -- will achieve a major breakthrough and its interaction patterns with humans will change significantly." According to 1AI, AI intelligences are artificial intelligences that can act independently and make decisions, and NVIDIA CEO Jen-Hsun Huang said in January this year that "the era of AI intelligences has arrived". And ...

April 1, 2011 - Amazon on Monday officially launched its general-purpose artificial intelligence agent, Nova Act, which has the ability to manipulate web browsers and autonomously perform simple web tasks. At the same time, Amazon also launched a supporting Nova Act software development kit (SDK), designed to help developers use Nova Act to build intelligent body application prototype. According to 1AI, Nova Act was developed by Amazon's newly established AGI (General Artificial Intelligence) Lab in San Francisco,...

I'm often asked how to build a commercial AI Agent. Should I choose Coze, Dify, or LangGraph? What are the considerations in the process? Where should I store my data? Why can't some web pages be crawled with this tool? ------ So, at the beginning of 2025, I started a collection, "Building an AI Agent from Scratch", combining my two years of experience in AI Agent development. The content contains the theory, practice and cases of AI Agent...

March 27, 2012 - Early this morning, OpenAI announced a major update to its Agent SDK, officially supporting Model Context Protocol (MCP) services. This technology allows developers to access unlimited third-party tools for intelligent bodies through a unified interface standard, significantly improving the development efficiency of complex automation applications. So, what is MCP? We can think of MCP as a "USB interface" for AI models, which provides a unified standard for AI intelligences to connect to...

A friend recently approached me and asked if I could come up with a tutorial on smart bodies, with a very strong willingness to pay! I really don't have the energy to make a set of detailed tutorials from 0 to 1. But I can see that this brother has a very good idea, but the technology is a short board, so I desperately want to learn how to build an AI intelligent body (because the AI intelligent body is really a fast track for ordinary people to quickly build a real working AI) If your intelligent body is done well, it can meet the needs of some people, and it can be used commercially to make money Although I don't have so much energy to engage in a detailed set of paid courses. But I'm still more than willing to share some free building AI...

March 13, 2011 - Microsoft co-founder Bill Gates, speaking on a podcast hosted by Reid Hoffman, looked ahead to a vision of the gradual integration of AI into every aspect of our lives, according to a report today in Windows Central. Gates said that the popularity of AI will be similar to when computers were first introduced. Today, most homes have at least one computer. Gates believes that AI will follow a similar trajectory, and eventually everyone will have a dedicated AI assistant to fulfill their personal needs. "In the future, each of us will...

March 6, 2011 - After the launch of Manus, a general-purpose Al Agent, sparked concerns in the tech industry, Manus AI partner Tao Zhang said today that he will focus on building in the short term. Tao Zhang also clarified some of the recent topics related to Manus. Thank you for your interest in Manus, and we would like to clarify that: we have never opened any channels to pay for invitation codes; we have never invested any marketing budget; the system capacity is limited during the beta period, so we will prioritize the core experience of our existing users and release invitations in an orderly manner. 1A...

Manus is a general-purpose AI Agent (Artificial Intelligence Agent) platform developed by a Chinese team to extend human capabilities and help users realize various visions. By connecting thoughts and actions, the platform not only thinks but also delivers results, excelling in a variety of tasks at work and in life, such as file processing, data analysis, code writing, content creation, etc., and can do everything while the user is resting. The design concept of Manus is "Knowing and Doing", based on intelligent means to expand human capabilities and become an intelligent partner of human beings. Currently, Manus is still in the internal testing stage, and needs to be used...

February 28 news, glory today said that it has cooperated with Ali in the field of AI to provide services for global users. At present, a variety of language, visual understanding, image generation models such as Qwen and Wan have been connected to the glory YOYO intelligent body, and users can experience AI functions through Magic7 series cell phones and other terminals. Prior to this, Glory YOYO assistant went online on February 19th with DeepSeek-R1 full-blooded version, which supports one-step switching of DeepSeek deep thinking mode. Glory phone YOY...

February 20 news, the first half of last year, a number of AI "empowered" gadgets have entered the market, but their respective fates are very different. The once-hot Humane AI Pin was "sentenced to death" yesterday, but Rabbit, which debuted at the same time, is still moving forward. Rabbit's blog post today shows a "universal Android smartbody" that slowly controls apps on a tablet in a similar way to what Rabbit claimed its R1 device could do a year ago, based on Rabbit's "universal smartbody" introduced last year. ...

Just now, Buckle officially announced that Buckle, for all users, has supported the use of DeepSeek R1/V3 models for a free experience. DeepSeek-R1 model The model will think deeply through the chain of thoughts before generating the final reply. Free to use, limited to 20 conversations/day per user. For more conversations, you can open Buckle Pro and create a DeepSeek-R1 model reasoning access point, billed by Token usage. DeepSeek-V3 Model The model has stronger language understanding and raw...

January 23, 2012 - Smart Spectrum announced today that GLM-PC, the "world's first" public-facing, turnkey computer intelligence, is open for experience, capable of "observing" and "operating" a computer like a human. GLM-PC is able to "observe" and "operate" the computer like a human being, and assist users to complete all kinds of computer tasks. According to the official introduction, GLM-PC by the "left brain" and "right brain" composition, "left brain" is responsible for rigorous logical reasoning and task execution, "right brain" is responsible for the depth of the computer tasks. The right brain" is responsible for depth perception and interactive experience. 1AI summary of the specific functions are as follows: GLM-PC can quickly develop a detailed task planning program according to the user's task requirements. It will synthesize...

ChatGPT has recently introduced a new feature called Tasks. It allows you to set up automated tasks to ensure that you receive reminders or responses from ChatGPT at a predetermined time. For example, you can set ChatGPT to send you a newsletter every morning, or automatically generate your daily task list, and even set reminders to get you into a good daily habit. What's more, once you set a task, ChatGPT will automatically send you a push notification or email reminder on time. So, this is ChatGPT to A...

The main content of this article: 1, AI Intelligent Body 2, what is the button and the basic concept of the button 3, the button intelligent system production process (as an example of microfilm) I. AI Intelligent Body AI Intelligent Body (AI Agent) is an intelligent system that can perceive the environment, analyze the information, and take action to achieve a specific goal. It combines artificial intelligence technology and is able to make decisions, learn and perform tasks autonomously. An AI Agent is similar to a virtual "assistant" or "agent". It can listen to you, understand your needs, and help you accomplish tasks. In the current big fire on the various intelligent body platforms, you can see...

January 4 news, Ali Tongyi Qwen launched a new AI framework Qwen-Agent, based on the existing Qwen language model, support for intelligent bodies to perform complex tasks, and provide a variety of advanced features to empower developers to build a more powerful AI intelligent body. The official introduction is attached as follows: Qwen-Agent is a development framework. Developers can develop Agent applications based on this framework to fully utilize the command following, tool usage, planning, and memory capabilities based on the Qwen model. This project also provides a browser assistant ...

January 1 afternoon news, Sina Technology was informed that Baidu recently in the Baidu search Web home page on-line Baidu "AI search" (Chat.baidu.com) entrance, to promote Baidu search revamped and upgraded. Sina Technology learned that the "AI search" based on the original Baidu search AI partner revamped and upgraded, on the basis of previous functional upgrades. Sina Technology learned that Baidu "AI Search" is based on Baidu Wenxin large model to build the desktop AI search engine, the content side has been through the Baidu search engine, Baidu Health, Baidu Lulin, Baidu library, Baidu education ...

Smart Spectrum Technology Team public yesterday (December 26) published a blog post announcing the open source of GLM-PC's base model CogAgent-9B-20241220, based on GLM-4V-9B training, dedicated to intelligent body (Agent) tasks. Note: The model only needs screenshots as input (no textual representation such as HTML), and can predict the next GUI operation based on any user-specified task, combined with historical operations. Thanks to the universality of screenshots and GUI operations, CogAgent can be widely used...

December 27 news, 2024 Lenovo Tianxi Eco-Partner Conference was held on December 26 at the Beijing International Convention Center. Lenovo, together with Intel, face wall intelligence, volcano engine and other companies, centered on the theme of "AI ecological future has come", and jointly discussed the development opportunities of Tianxi ecological development and Tianxi AS technology trends. At the conference, the new generation of intelligent body system Tianxi Personal Intelligent Body System (Tianxi AS) was upgraded. The conference also witnessed the inauguration of the AI Terminal Intelligent Body Innovation Professional Committee of the China Software Industry Association. In 2024, Lenovo fully promotes...

Recently, the button has a major update, the introduction of a new AI application form, the original some of the image flow function has a large update changes, the old tutorials are no longer applicable, the new features have become simpler and more powerful. Some friends asked how to make the original Portrait Portrait Intelligent Body. Today, the revamped features to re-create a replacement for the face of the portrait of the smart body, for the need of students to update the study. First look at the effect Left: the user uploads a picture of the face Middle: the template base image Right: the user uploads a picture + template image generated by the new personal portrait effect. Below, learn together how to make a replacement face with a button...

According to a report by @GuoYiTongZhong on December 19, the first domestic intelligent body security ecological collaboration organization - "IIFAA Intelligent Body Trusted Interconnection Working Group" (hereinafter referred to as the "Working Group") announced the establishment of the Working Group in Beijing. The working group is dedicated to promoting the industry to jointly develop relevant security technical specifications for cross-smart body interactions, in order to respond to the security and collaboration challenges brought about by the rapid development of smart bodies, and to jointly build an open, cooperative, and win-win smart body ecosystem. 1AI learned from the report that more than twenty head of the China Academy of Information and Communications Technology, Glory, OPPO, Xiaomi, Huawei, Smart Spectrum, Byte, Ali, Ant, Jingdong, China Mobile, Qualcomm and other...