Model Anatomy See what's inside open-weight AI models

Can I run Huihui-Qwen3.8-27B-abliterated-GGUF on a Mac with 36 GB?

Yes, just

Its Q5_K_L file (22.9 GB) fits in the 24.0 GB a Mac with 36 GB can use, with room for ≈ 15K tokens of context.

Full analysis of Huihui-Qwen3.8-27B-abliterated-GGUF Everything that runs on Mac with 36 GB Get the Q5_K_L file on Hugging Face
ollama run hf.co/huihui-ai/Huihui-Qwen3.8-27B-abliterated-GGUF:Q5_K_L

Files and context on Mac with 36 GB

Each file of Huihui-Qwen3.8-27B-abliterated-GGUF (and of GGUF versions the site has analyzed), and how much context fits next to it, with the context cache in 16-bit or 8-bit.

FileSizeOn Mac with 36 GB Context (16-bit)Context (8-bit)
BF1654.7 GBDoesn't fit——
Huihui-Qwen3.8-27B-abliterated-bf1654.7 GBDoesn't fit——
F1653.8 GBDoesn't fit——
Huihui-Qwen3.8-27B-abliterated-Q8_0_L38.8 GBDoesn't fit——
UD-Q8_K_XL31.5 GBDoesn't fit——
Q8_029.0 GBDoesn't fit——
Q8_K_L27.3 GBDoesn't fit——
Q6_K_L24.9 GBDoesn't fit——
UD-Q6_K_XL24.8 GBDoesn't fit——
Huihui-Qwen3.8-27B-abliterated-UD-DW-Q6_K_L23.9 GBDoesn't fit——
Q5_K_L22.9 GBTight15K29K
Q6_K22.4 GBTight21K42K
Q4_K_L20.9 GBTight44K87K
UD-Q5_K_XL20.7 GBTight47K95K
Q5_K19.5 GBTight64K128K
Q3_K_L18.7 GBFits77K154K
UD-Q4_K_XL17.4 GBFits96K193K
Q2_K_L17.2 GBFits99K197K
Q4_K16.8 GBFits105K210K
Q4_K_M16.6 GBFits109K217K
Q4_K_S15.6 GBFits123K245K
UD-IQ4_XS14.4 GBFits141K256K
Q3_K13.5 GBFits154K256K
UD-Q3_K_XL13.3 GBFits157K256K
Huihui-Qwen3.8-27B-abliterated-UD-DW-IQ3_S12.3 GBFits172K256K
Huihui-Qwen3.8-27B-abliterated-Swift-1.5-GSQ-RCO-IQ3_S-mtp12.1 GBFits175K256K
IQ3_S12.1 GBFits175K256K
UD-IQ3_S12.0 GBFits177K256K
Huihui-Qwen3.8-27B-abliterated-Swift-1.5-GSQ-RCO-IQ3_S11.8 GBFits180K256K
Huihui-Qwen3.8-27B-abliterated-GSQ-RCO-IQ3_S11.8 GBFits180K256K
Huihui-Qwen3.8-27B-abliterated-UD-DW-IQ3_XXS11.4 GBFits186K256K
UD-IQ3_XXS11.0 GBFits191K256K
Q2_K10.9 GBFits193K256K
Huihui-Qwen3.8-27B-abliterated-Swift-1.5-GSQ-RCO-IQ3_XXS-mtp10.4 GBFits200K256K
IQ3_XXS10.4 GBFits200K256K
Huihui-Qwen3.8-27B-abliterated-Swift-1.5-GSQ-RCO-IQ3_XXS10.1 GBFits205K256K
Huihui-Qwen3.8-27B-abliterated-GSQ-RCO-IQ3_XXS10.1 GBFits205K256K
UD-Q2_K_XL10.0 GBFits206K256K
IQ2_S9.61 GBFits212K256K
Huihui-Qwen3.8-27B-abliterated-Swift-1.5-GSQ-RCO-IQ2_S9.26 GBFits217K256K
UD-IQ2_S8.78 GBFits225K256K
IQ2_XS8.76 GBFits225K256K
Huihui-Qwen3.8-27B-abliterated-UD-IQ2_S8.42 GBFits230K256K
Huihui-Qwen3.8-27B-abliterated-Swift-1.5-GSQ-RCO-IQ2_XS8.42 GBFits230K256K
Huihui-Qwen3.8-27B-abliterated-Underdog-Saluki-Swift-1.5-IQ2-mtp8.25 GBFits232K256K
Huihui-Qwen3.8-27B-abliterated-Underdog-Saluki-Swift-1.5-IQ27.90 GBFits238K256K
Huihui-Qwen3.8-27B-abliterated-Ternary-Bonsai-PQ2_07.70 GBFits241K256K
Huihui-Qwen3.8-27B-abliterated-Ternary-Bonsai-PTQ1_06.61 GBFits256K256K

Huihui-Qwen3.8-27B-abliterated-GGUF on other hardware

Popular models that run on Mac with 36 GB

Assumes 67% of the Mac's memory is usable (macOS's default limit for the GPU; it can be raised) and at least 4K tokens of context for one conversation, with the context cache in 16-bit unless noted. Fits leaves at least 20% of that memory for context; Tight leaves less.