Xiaomi open-sourced MiMo-V2.6-Pro, the highest-scoring open-source model on the…
This is a AI post classified by Jev as Open & local models (a model release), kept by the AI Radar because it carries real work, not commentary.
Xiaomi open-sourced MiMo-V2.6-Pro, the highest-scoring open-source model on the Artificial Analysis Intelligence Index. 🔥 It scores 46.32, ahead of Kimi K3 and Qwen3.8 Max. On DeepSWE v1.1, a held-out long-horizon software engineering benchmark, Pro moved from 58.4 to 72.57. Pro also wrote Night Road, an orchestral score for about ten instruments, and converted it to MIDI on its own. It helped formalize the main theorem in Li and Yorke's Period Three Implies Chaos. After further revision, the Lean 4 proof was more than 6,000 lines, and the kernel verified it. The model had no Lean-specific
Posted by Markets & Mayhem (271.3k followers) 2 h ago · 11 likes · 5.3k views · view the original post on X. Kept by the AI Radar as Open & local models.
More AI work like this
- Meet decider-2b: a 2B parameter decision model that runs in ONE pass. No chains, no… — @HuggingModels
- Xiaomi’s MiMo V2.6 Pro looks like a contender for the best new model to run locally. — @plotarmordev
- Here's a model that shouldn't be overlooked! — @mr_r0b0t
- I let Kev play Dwarf Fortress — @TelepathicPug
- llama.cppで起動したMiMo-V2.6-Distill-Qwen-9Bにやらせた。ラーメン屋ベンチの結果です。動画をご査収ください。 — @studio_yebisu
- the open frontier intelligence list keeps on growing. — @arjunkocher
- Qwen Image 2.1 Is OPEN WEIGHTS… Nano Banana 2 at HOME ! https://youtu.be/6teCD4Kq4wA — @Tech2Wild
- Yesterday you could already make realistic images locally with Krea 2 on an RTX card. — @yume_arasaki
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 40.6k posts from 5k X accounts over the last 14 days, 1.7k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-22 00:34 UTC. Full method.