216 3 weeks ago

Quantized Q3_K_M version of Qwen3.6 27b delivers substantial upgrades in agentic coding and thinking preservation than previous Qwen models.

ollama run siendsi/qwen3.6-27b-q3km-256k

Details

3 weeks ago

6be6ba64a5b9 · 14GB ·

qwen35
·
27.3B
·
Q3_K_M
{ "num_ctx": 262144 }

Readme

The original model is here https://ollama.com/library/qwen3.6 is completely consistent with the original, the quanting damage is not significant.