%20--%3e%3cdefs%3e%3cstyle%3e%20.st0%20{%20fill:%20%23061b40;%20}%20.st1%20{%20fill:%20%23306af1;%20}%20.st2%20{%20fill:%20%235ce5cf;%20}%20%3c/style%3e%3c/defs%3e%3cg%3e%3cpath%20class='st0'%20d='M55,10.5h9v3h-9v9h12V7.5h-12v3ZM64,19.5h-6v-3h6v3Z'/%3e%3cpolygon%20class='st0'%20points='69%2016.5%2078%2016.5%2078%2019.5%2069%2019.5%2069%2022.5%2081%2022.5%2081%2013.5%2072%2013.5%2072%2010.5%2081%2010.5%2081%207.5%2069%207.5%2069%2016.5'/%3e%3cpolygon%20class='st0'%20points='95%2010.5%2095%207.5%2083%207.5%2083%2022.5%2095%2022.5%2095%2019.5%2086%2019.5%2086%2016.5%2095%2016.5%2095%2013.5%2086%2013.5%2086%2010.5%2095%2010.5'/%3e%3cpath%20class='st0'%20d='M40,1.5v21h11.6l1.4-1.4v-7.6h0c0,0-1.4-1.5-1.4-1.5l1.4-1.4V2.9l-1.4-1.4h-11.6ZM50,19.5h-7v-6h7v6ZM50,10.5h-7v-6h7v6Z'/%3e%3c/g%3e%3cpath%20class='st1'%20d='M23.1,24L14.7,4.8l-4.9,11.2h3.8l-1.8,4H3.3L12.1,0H2C.9,0,0,.9,0,2v20c0,1.1.9,2,2,2h21.1Z'/%3e%3cpath%20class='st2'%20d='M34,0h-16.8l10.6,24h6.2c1.1,0,2-.9,2-2V2C36,.9,35.1,0,34,0ZM32.5,20h-4V4h4v16Z'/%3e%3c/svg%3e)
Today, the Xiaomi technology team officially released and open-sourced an exploratory AI programming assistant running on the terminal - MiMo Code V0.1.0. This project is a secondary development based on the open-source project OpenCode, and it is open-sourced under the permissive MIT license, allowing individuals and companies to freely modify, distribute, and integrate commercially.
As a self-evolving tool that focuses on "model Agent collaborative optimization," MiMo Code aims to break the limitations of traditional AI coding tools. Its core highlights include:
Innovative persistent memory system, solving the "AI forgetfulness": Traditional tools rely on the model's own awareness of recording, which can easily lose key information during long conversations. MiMo Code introduces three mechanisms: project memory, session checkpoints, and task progress, fully "outsourcing" the recording work to an independent Subagent. When the window is full, the Subagent automatically rebuilds a clean summary, allowing the main Agent to seamlessly continue. In addition, the built-in
/dreamcommand triggers automatically every 7 days to deduplicate, verify, and compress historical sessions, achieving memory consolidation, making the tool "understand you better over time."Dedicated Harness system and Compose mode: To address the low compatibility between models and frameworks, Xiaomi has developed a dedicated Harness system for its built-in model. Users just need to press the Tab key to switch to Compose mode, input a simple idea, and the AI will automatically complete the entire process of "design-planning-coding-testing-review." In real-world testing comparisons with Claude Code, MiMo Code adopts a strategy of "heavy planning upfront and stable verification later," which is slightly slower at first but delivers more feature-rich, well-tested industrial-grade products, resulting in less overall time spent.
Excellent performance on authoritative test sets: With the same MiMo model, relying solely on the collaborative advantages of the Agent system, MiMo Code achieved a score of 62% on SWE-Bench Pro (leading Claude Code's 57%), and 73% on Terminal Bench2 (leading Claude Code's 68%).
Support for top-tier models and full Chinese localization: The tool includes a limited-time free multimodal model MiMo-V2.5 (performance comparable to Claude Sonnet4.6), which can be used out of the box without registration. It also supports mainstream APIs and third-party Token Plans from DeepSeek, Kimi, GLM, etc. A status dashboard is always displayed on the right side of the interface, and all settings have been fully translated into Chinese.
Voice interaction without manual operation: Relying on MiMo-V2.5-ASR voice recognition technology, users not only can modify instructions verbally, but also directly say commands like "send" and "execute" to achieve full verbal control.
Blog:https://mimo.xiaomi.com/mimocode