Inside LFM2.5-2.6B's post-training recipe: SFT, RL, multi-domain distillation
helloiamleonie · x · 2026-09-25
Leonie assembles the full post-training recipe for LFM2.5-2.6B: (1) classical SFT with agent traces and antidoom training; (2) SFT+RL to train specialized teachers per skill; (3) MOPD distills all skills back into one checkpoint; (4) agentic RL polishes the model for agent harnesses. The 2.6B model competes with the 3x larger Qwen3.5-9B.
Related event: Liquid AI Releases LFM2.5-2.6B and Details Its On-Device Training Recipe(10 posts)→
More from Models
- Google criticized for locking $20 Workspace AI subscribers to outdated Gemini models — thedealdirector · 2026-09-25
- Google's $20 Workspace AI add-on slammed for leaving enterprise Gemini stuck on old Flash model — thedealdirector · 2026-09-25
- Hands-on with StepFun Step-5-Preview: rock-solid agent loops, weak on 3D and aesthetics — karminski3 · 2026-09-25
- Claude Opus 5.5 scores 31.2% on WeirdML v3, trailing GPT 6 Astra's 42.2% — scaling01 · 2026-09-25
- Sora API shuts down today with no replacement from OpenAI — VraserX · 2026-09-25
- 'LLMs will eventually write assembly' — Greg Mushen finally saw an example that convinced him — gregmushen · 2026-09-25