Manus
Manus is an AI agent product capable of autonomously planning and executing complex tasks, including invoking tools, operating computer environments, and executing code to accomplish user-specified objectives. Its core characteristics include multi-step task automation, tool-calling capability, and a high degree of autonomous decision-making, designed to reduce manual intervention and improve the execution efficiency of complex workflows.
Core Facts
Timeline (last 90 days)
The AI agent market currently already has multiple products such as Viktor, Lindy, Manus, OpenClaw, Microsoft Copilot Cowork, and Zapier AI agents, and the field is crowded.
General-purpose AI colleagues such as Manus and Microsoft Copilot Cowork emphasize completing tasks end-to-end, but users report that their reliability is unstable and that complex tasks easily deviate from expectations.
From Manus, which exploded in popularity early this year, to the recently viral Cos and DeFi, and then to DeepSeek, behind them all is the same logic of turning large models into intelligent agents that can get work done.
Manus survived after Meta's $2 billion acquisition was halted by China, and raised another $500 million in October.
In the agent sector, doing only the orchestration layer makes for a shallow moat and may be covered by the native agent capabilities of foundation model vendors, so developing the foundation upstream is an industry trend.
Chinese agent company Manus has just completed over $500 million in financing at a $4 billion valuation and has officially started developing its own underlying large model, shifting from orchestrating external models to controlling its core foundation.
Manus has been reported to be training its own large model, shifting from calling external models such as Claude and Qwen to in-house development, has raised over $500 million, and surpassed $100 million in annualized recurring revenue 8 months after launch.
Manus Game Dev is positioned as a rapid prototyping tool rather than a full game engine, and its optimization target is the path from idea to playable demo, not replacing Unity's or Unreal's production pipelines.
Manus launched a game generation feature called Game Dev that can automatically handle 3D scenes, character rigging and animation construction, directly generating games that run in the browser.
Agents can be divided into three categories: workflow-based agents (based on DAG orchestration, such as Coze/Dify), general-purpose agents (relying on planning capabilities to autonomously decompose tasks, such as Manus), and vertical agents (targeting specific domains, such as CapCut's Smart Editing and Figma AI).
31 more timeline events
All Facts (20)
Manus is a general-purpose AI Agent product launched by a Chinese team in early 2025.
80%VerifiedManus AI is built on a Cloud VM architecture that simulates a complete computer environment on remote servers to execute tasks.
65%UnverifiedYuManus是对AI Agent框架Manus的Java实现，采用ReAct（Reasoning + Acting）模式
90%UnverifiedManus于2025年3月初首次亮相
90%UnverifiedManus在2025年初发布的演示视频展示了自动完成市场调研、生成完整报告等复杂工作流的能力
85%UnverifiedManus并没有自研底层大模型，本质上是'套壳'他人模型
85%UnverifiedManus在GAIA基准测试中取得了领先成绩
80%UnverifiedManus、Devin等主流AI Agent平台普遍采用Docker容器隔离的架构方案
80%UnverifiedManus的邀请码一度在二手市场被炒到数百元一个
75%UnverifiedCodeAct论文的推动者是Manus的联合创始人
70%UnverifiedManus等新一代Agent能通过代码执行系统命令，存在误删文件、修改配置等安全风险
70%UnverifiedManus核心AI技术可能属于《中国禁止出口限制出口技术目录》中的信息处理技术
70%UnverifiedManus发布后邀请码在二手平台被炒到10万元一个，两天内微信指数超越DeepSeek
70%UnverifiedManus's technical architecture adopts a multi-Agent collaborative design, with multiple specialized sub-Agents (search Agent, analysis Agent, writing Agent, code execution Agent) working together to complete tasks
60%UnverifiedManus's Series B funding was reviewed by the U.S. Treasury Department's Outbound Investment Security Program, and investors including Benchmark required Manus to relocate its headquarters out of China.
60%UnverifiedIn April 2025, Manus completed a $75 million Series B funding round led by Silicon Valley's Benchmark, with its valuation surging from $85 million to $500 million.
60%UnverifiedThe AI agent market currently already has multiple products such as Viktor, Lindy, Manus, OpenClaw, Microsoft Copilot Cowork, and Zapier AI agents, and the field is crowded.
50%UnverifiedGeneral-purpose AI colleagues such as Manus and Microsoft Copilot Cowork emphasize completing tasks end-to-end, but users report that their reliability is unstable and that complex tasks easily deviate from expectations.
50%UnverifiedFrom Manus, which exploded in popularity early this year, to the recently viral Cos and DeFi, and then to DeepSeek, behind them all is the same logic of turning large models into intelligent agents that can get work done.
50%UnverifiedIn the agent sector, doing only the orchestration layer makes for a shallow moat and may be covered by the native agent capabilities of foundation model vendors, so developing the foundation upstream is an industry trend.
50%