atomic[.]chat (a desktop app that runs LLMs locally) ran a very revealing comparison for local AI agents, on a MacBook Pro M5 Max, 64GB. Liquid’s much smaller LFM2.5-8B-A1B beat gpt-oss-20b by…

51 tokens/sec just became 117 on the same hardware 👀 Atomic Chat just added Multi-Token Prediction (MTP) for Qwen models. • Qwen3.6 27B → 51 → 117 tps (+137%) • Qwen3.6 35B-A3B → 218 → 267 tps…

Qwen 3.6 models are now 2.5x times faster on Atomic Chat with new MTP speedups. > MTP drafts several tokens ahead and verifies them in one pass. The speedup depends on the memory moved per pass.…

🧵 1/2 Tired of the endless setup dance for private LLMs? Atomic Chat just dropped v1.1.76, an optimized inference stack that lets you run large language models at near‑desktop speed on any…

Another good news for local-LLM from atomic[.]chat, that runs 100% offline on your computer. They just showed MTP (Multi-Token Prediction) pushing local Qwen models from 51 to 117 tokens/s on dense…