Hugging Face launches live training streams, starting with RL-tuned LoRA adapters
art_zucker · x · 2026-11-08
Hugging Face now lets anyone watch live training runs on the platform.
The first showcase is Harness adapters training: LoRA adapters are being trained with RL on LiquidAI's LFM2.5 / LFM3 and the recent Twil-9B model, aiming to make them more efficient and accurate on agentic tasks — with the whole process streamable in real time.
More from Infra
- Running 50-100 Agents, Dev Weighs Local Decision Models to Cut Latency — f1yingbanana · 2026-11-08
- TSMC denies 3D DRAM plans at Nvidia's request, stays pure-play logic foundry — rohanpaul_ai · 2026-11-08
- TSMC denies developing 3D DRAM at Nvidia's request, staying a pure logic foundry — rohanpaul_ai · 2026-11-08
- Why is everyone sleeping on Modal live inference endpoints for agents? — vboykis · 2026-11-08
- Arista CEO forecasts 10+ GW of AI deployments across 10+ accelerator types by end of 2027 — Beth_Kindig · 2026-11-08
- Lobsters thread: how AI traffic is degrading the web for human developers — vboykis · 2026-11-08