webAI releases open-source 3.6B local reasoning model TwIL-LM3-Pro

webAI has released TwIL-LM3-Pro, a 3.66B open-source model for local reasoning whose Q4 quantized version is only 2.09GiB and runs via llama.cpp on CPU or local GPU. The weights are now available on Hugging Face.

2026-10-01 ~ 2026-10-01 · 2 related posts

• webAI's 3.6B TwIL-LM3-Pro beats VibeThinker-3B by 35% on formal logic, runs locally in 2GiB — rohanpaul_ai · 2026-10-01
• TwIL-LM3-Pro weights now live on Hugging Face — rohanpaul_ai · 2026-10-01