A phone company just took the open-weight crown. xiaomi’s mimo-v2.6-pro tops benchlm’s october…
a phone company just took the open-weight crown. xiaomi’s mimo-v2.6-pro tops benchlm’s october ranking at 75.5, ahead of qwen3.8 max at 72.1, and the release includes 7,000 rl task environments under mit, not just weights.
the training recipe that used to be the private moat is the part they gave away.
Context
Verified source: Xiaomi's MiMo page says the MiMo-V2.6 series has two native multimodal models, Pro and Flash, and that MiMo-V2.6-Pro scores 46 on the Artificial Analysis Intelligence Index, ahead of Kimi K3 and Qwen3.8 Max among open-source models. BenchLM's model page, dated Sep 22, lists it as an open-weight model with a 1M context window.
The Hugging Face card describes MiMo-V2.6-Pro-RL as the flagship checkpoint and an arXiv paper, MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement, describes the RL approach.
BenchLM's page read here gives MiMo-V2.6-Pro 74.15 out of 100, ranked #11 of 889, and not the 75.5 in the post; the October ranking the post cites was not seen in the sources read, so unsupported here, not refuted. The 72.1 for Qwen3.8 Max and the 7,000 RL task environments under MIT were also not seen in the sources read, so unsupported here, not refuted. Xiaomi's own claim of the top open-source spot rests on the Artificial Analysis index, a different benchmark from BenchLM. 'The recipe that used to be a moat was given away' is the author's opinion.
Related work
- MiMo-V2.6-Pro-RL model card ↗Hugging Face card.
- BenchLM page ↗Sep 22 snapshot.
- Xiaomi MiMo news ↗Release post.
Watch next
- Read the arXiv paper for the environment count and the license terms of the release.
Sources
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 1 October 2026 at 07:38 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →