Xiaomi's model jumps near the frontier with just 30 RL steps, no plateau in sight
teortaxesTex · x · 2026-09-22
Commenting on Xiaomi's training details, nrehiew and others are surprised the model improved dramatically with only 30 RL steps — far fewer than expected — and question why training stopped given no visible plateau.
More from Models
- Datalab claims parsing accuracy second only to Gemini at 1/37 the cost and 6x speed — VikParuchuri · 2026-11-12
- Datalab fast-mode OCR beats Opus 5.5 and GPT-6.1 Sol at 1/4 to 1/68 the price — VikParuchuri · 2026-11-12
- Claude's world knowledge impresses: model guesses user identity from one prompt — adonis_singh · 2026-11-12
- Gemini 4 Carbon opens to all at $2/$10, price doubling to $4/$20 on Dec 1 — socialwithaayan · 2026-11-12
- Ten wild Gemini 4 Carbon demos in 24 hours: CUDA meadows, V8 engines, mech stomps — socialwithaayan · 2026-11-12
- How to cut Fable 5.2 token costs by 90%: Arc routing plus a lean CLAUDE.md — EXM7777 · 2026-11-12