Xiaomi MiMo-V2.5 Series: Pushing Open-Source Agents Forward
🔸 MiMo-V2.5-Pro, our strongest model yet.
A major leap from MiMo-V2-Pro in general agentic capabilities, complex software engineering, and long-horizon tasks, now matching frontier models like Claude Opus 4.6 and GPT-5.4 across most benchmarks (SWE-bench Pro 57.2, Claw-Eval 63.8, τ3-Bench 72.9). It can autonomously complete professional tasks involving 1,000+ tool calls, work that would take human experts days.
Tech Blog: https://t.co/I275kxl9kL
🔸 MiMo-V2.5, native omnimodal with strong agentic capabilities.
Pro-level agent performance at roughly half the cost. Improved multimodal perception across image and video understanding, native 1M-token context window, and significantly more efficient inference.
Tech Blog: https://t.co/7DLo0n2MNu
🔗 API & Token Plan:
The MiMo-V2.5 series is tuned for token efficiency.
At the same ClawEval agent benchmark score:
🔸 MiMo-V2.5-Pro uses 42% fewer tokens than Kimi K2.6
🔸 MiMo-V2.5 uses nearly half the tokens of Muse Spark
Same frontier performance, at a fraction of the cost.
The MiMo-V2.5 series — how they fit together:
🔸 MiMo-V2.5-Pro: built for long, complex agent tasks that demand deep reasoning.
🔸 MiMo-V2.5: covers most general-purpose agent scenarios at 100–150 tokens/s, with native omni-modal capabilities across image, audio, and video.
V2.5-Pro for the hardest tasks. V2.5 for speed and modality.
Token Plan Update
Alongside stronger models, we've upgraded our inference infrastructure — and your Token Plan gets better too.
Simpler, lower credit rates:
• MiMo-V2.5: 1x (1 token = 1 credit)
• MiMo-V2.5-Pro: 2x (1 token = 2 credits)
From now on, Token Plans (https://t.co/P74SR2MaJe) no longer charge a multiplier for the 1M context window.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
SWE-bench — The standard benchmark for AI coding agents: real GitHub issues from real repositories, scored by whether the agent's patch passes the project's own tests.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
Why it matters
Token efficiency at equal benchmarkA standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.Full definition → score is what decides AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → run cost. Xiaomi claims 42 percent fewer tokens than Kimi K2.6 at the same ClawEval score, with the 1M context windowThe maximum amount of text a model can consider at once — its working memory for the current conversation or task.Full definition → multiplier removed from plan billing.
postHeads up, agent users！ If you're using Xiaomi MiMo with thinking mode: When…
postMiMo-V2.5 and V2.5-Pro go open weights under MIT with day-zero SGLang and vLLM
postIntroducing MiMo-V2.5 Voice — our full-stack voice lineup for the Agent era. 🚀…
