Xiaomi's MiMo-V2.6 RL run: 2B tokens per step, live dashboard with running costs
burny_tech · x · 2026-09-17
Xiaomi's @LuoFuli says the team spent six months studying how far RL can scale; MiMo-V2.6 is mid-run now. They scaled compute (2B tokens/step, 1568 prompts × 16 rollouts, fully async), environments/harnesses (multi-task agentic RL mixing harnesses), and grader compute (in-group credit assignment with test-case and rubric rewards). Details will be open-sourced gradually; the run is streamed live with a public cost dashboard. giffmana quips all three 'things we scaled' are really just compute, but praises the transparent live cost panel.
Related event: Xiaomi live-streams MiMo-V2.6 RL training, burning over $1.08M in 36 hours(14 posts)→
More from Models
- Anthropic's server-side rule reignites the case for owning open models over renting AI — alex_verem · 2026-11-02
- Epoch AI: open-weight GLM 5.3 exploited the patch gap to hack 6 Firefox releases at ~$1,000 each — rohanpaul_ai · 2026-11-02
- SemiAnalysis: safety classifiers sandbag scores of Opus 5.5 and GPT-6 Astra — morqon · 2026-11-02
- Liquid AI's 1.7B LFM3 becomes first local model to nail live fixtures lookup on a MacBook — NielsRogge · 2026-11-02
- Claude upgrade-page traffic rebounded to 17.6M in September, near May's 18.6M peak — firstadopter · 2026-11-02
- Open-source GLiNER-PII scrubs 13 identifiers from a discharge note in 550ms on GPU — MaziyarPanahi · 2026-11-02