š MiMo Code V0.1 is now live and open-sourceļ¼ More than an AI coding assistantā¦
- Source
- x.com
- Date
š MiMo Code V0.1 is now live and open-sourceļ¼ More than an AI coding assistant in your terminal ā it's the smartest coding partner you'll ever work with. Comes with MiMo V2.5, a multimodal model available free for a limited time, featuring a million-token context windowāready to use out of the box. ā¾ļø Infinite Context: Knowledge accumulates automatically, and with lossless compression, even million-line projects keep every critical detail intactāquality never drops. š§ Agent-Model Synergy: An Agent framework deeply optimized for MiMo, with a full closed loop of testing, review, and validationāso complex tasks get done in one pass. š Compose Mode: Specs ā Plans ā Build ā Report. Design first, code secondāclear thinking, no rework. š Self-Evolving System: Every session is automatically reviewed, distilling experience and best practicesāthe more you use it, the smarter it gets. šļø Voice Input: Powered by MiMo-V2.5-ASR ā just speak instead of type, and your voice becomes the prompt for truly hands-free coding. š Claude Code Compatible: Automatically loads your existing skills, MCP servers and commands, and reuses your API configurationāzero-cost migration, no setup required.ā¦
Context
Xiaomi's MiMo lab released MiMo Code V0.1 as an MIT-licensed, open-source terminal coding bundled with MiMo V2.5, a model the company says supports a million-token . Xiaomi describes the CLI as automatically loading a user's existing Claude Code skills, servers and commands, and reusing existing API configuration rather than requiring separate setup.
The company also describes a "Compose Mode" that separates a specification-and-plan step from the build step, an automatic session-review loop it calls self-evolving, and voice input powered by a Xiaomi speech-recognition model. The CLI supports switching between model providers beyond Xiaomi's own, including Anthropic, OpenAI, DeepSeek, Kimi and GLM, according to the announcement.
- AI agent ā An AI system that doesn't just answer once but works toward a goal in a loop ā taking actions, reading the results, and deciding what to do next.
- multimodal ā A model that works with more than text ā reading images, audio, or video, and sometimes generating them too.
- context window ā The maximum amount of text a model can consider at once ā its working memory for the current conversation or task.
- MCP ā The Model Context Protocol ā an open standard that lets any AI assistant plug into any tool or data source without custom integration code.
postXiaomi Releases MiMo-V2.6, an Open-Weight Model Claiming Top Benchmark Score- postXiaomi opens up the inference stack behind MiMo-V2.5
postHeads up, agent usersļ¼ If you're using Xiaomi MiMo with thinking mode: Whenā¦
postMiMo-V2.5 and V2.5-Pro go open weights under MIT with day-zero SGLang and vLLM
Checking sign-inā¦
Loading commentsā¦