Qwen3.6-35B-A3B

Latest news

1. 15 Oct
   Hugging Face releases LFM2.5-24B-A2B for fast on‑device agents

   Hugging Face announced LFM2.5-24B-A2B, a 24B-parameter mixture-of-experts reasoning model that activates only 2B parameters per token to run fast, private agents on local hardware. The model is open-weight, competitive with larger models on agentic benchmarks, and is available on Hugging Face and the company’s Playground.

   Models · 2 sources

   

Articles

No articles about Qwen3.6-35B-A3B yet.