Xiaomi MiMo V2.6 Pro tops open-source leaderboard at 46, costs ~$0.13 per task
heyshrutimishra · x · 2026-09-23
Per Artificial Analysis, Xiaomi's MiMo V2.6 Pro is now the #1 open-source model globally and #1 among Chinese models, scoring 46 on the Intelligence Index — about one point behind GPT-5.6 Sol — at a weighted cost of roughly $0.13 per task.
- The new lineup spans Pro, UltraSpeed and Flash; flagship Pro is fully multimodal, targeting complex projects, long-horizon work, cybersecurity and research.
- vs MiMo V2.5, intelligence nearly doubles at unchanged prices, driven by post-training around Environment × Task × Grader with no pre-training changes.
More from Models
- Opus 5.5 feels nerfed — user blames system prompt restrictions, not the model — oyacaro · 2026-11-12
- ChatGPT now links to brand homepages far more, reshaping brand-mention tracking — lilyraynyc · 2026-11-12
- YC-backed Monte launches post-training platform to build proprietary models from your agents' real work — ycombinator · 2026-11-12
- Leak Claims OpenAI Will Launch Entire GPT-6.2 Family Tomorrow to Counter Opus 5.5 and Fable 5.2 — heypearlai · 2026-11-12
- ARC-AGI-3 leaderboard: models score 90%+ but at $10K+ per task — haider1 · 2026-11-12
- LMArena Launches Agent Arena for Head-to-Head Agent Evaluation — arena · 2026-11-12