Request iq3_m or q3_k_m quants please
#7
by Neoder - opened
Hi! Thank you for sharing this model.
Could you please consider adding lower-bit GGUF quants, specifically iq3_m or q3_k_m?
Both are up now — Q3_K_M and IQ3_M, each with the MTP head baked in like the rest of the ladder:
ThinkingCap-Qwen3.6-27B-Q3_K_M-MTP.gguf 12.6 GiB
ThinkingCap-Qwen3.6-27B-IQ3_M-MTP.gguf 11.9 GiB ← smallest full-quality rung, fits a 16 GB card
artificial-citizen changed discussion status to closed