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

Can I run Arete-27B on a Mac with 24 GB?

Yes, with a GGUF version

The published weights are too big, but the IQ4_XS file from bartowski/NousResearch_Arete-27B-GGUF (15.2 GB) fits, with room for ≈ 9K tokens of context.

Full analysis of Arete-27B Everything that runs on Mac with 24 GB Get the IQ4_XS file on Hugging Face
ollama run hf.co/bartowski/NousResearch_Arete-27B-GGUF:IQ4_XS

Files and context on Mac with 24 GB

Each file of Arete-27B (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 24 GB Context (16-bit)Context (8-bit)
As published54.7 GBDoesn't fit——
BF16bartowski/NousResearch_Arete-27B-GGUF53.8 GBDoesn't fit——
Q8_0bartowski/NousResearch_Arete-27B-GGUF28.7 GBDoesn't fit——
Q6_K_Lbartowski/NousResearch_Arete-27B-GGUF24.7 GBDoesn't fit——
Q6_Kbartowski/NousResearch_Arete-27B-GGUF23.6 GBDoesn't fit——
Q6_K_Sbartowski/NousResearch_Arete-27B-GGUF22.6 GBDoesn't fit——
Q5_K_Mbartowski/NousResearch_Arete-27B-GGUF20.7 GBDoesn't fit——
Q5_K_Sbartowski/NousResearch_Arete-27B-GGUF19.3 GBDoesn't fit——
Q4_K_Lbartowski/NousResearch_Arete-27B-GGUF18.6 GBDoesn't fit——
Q4_1bartowski/NousResearch_Arete-27B-GGUF17.6 GBDoesn't fit——
IQ4_NLbartowski/NousResearch_Arete-27B-GGUF17.2 GBDoesn't fit——
Q4_K_Mbartowski/NousResearch_Arete-27B-GGUF17.2 GBDoesn't fit——
Q4_K_Sbartowski/NousResearch_Arete-27B-GGUF16.1 GBDoesn't fit——
Q4_0bartowski/NousResearch_Arete-27B-GGUF16.1 GBDoesn't fit——
IQ4_XSbartowski/NousResearch_Arete-27B-GGUF15.2 GBTight9K18K
IQ3_Mbartowski/NousResearch_Arete-27B-GGUF14.6 GBTight18K36K
Q3_K_Lbartowski/NousResearch_Arete-27B-GGUF13.9 GBTight29K58K
Q3_K_Mbartowski/NousResearch_Arete-27B-GGUF13.2 GBTight40K80K
IQ3_XSbartowski/NousResearch_Arete-27B-GGUF12.6 GBFits49K98K
Q3_K_Sbartowski/NousResearch_Arete-27B-GGUF12.5 GBFits50K100K
IQ3_XXSbartowski/NousResearch_Arete-27B-GGUF12.1 GBFits56K112K
Q2_Kbartowski/NousResearch_Arete-27B-GGUF10.6 GBFits78K157K
IQ2_Mbartowski/NousResearch_Arete-27B-GGUF10.3 GBFits83K166K
IQ2_Sbartowski/NousResearch_Arete-27B-GGUF9.45 GBFits95K191K
IQ2_XSbartowski/NousResearch_Arete-27B-GGUF8.85 GBFits104K209K
IQ2_XXSbartowski/NousResearch_Arete-27B-GGUF8.64 GBFits107K215K

Arete-27B on other hardware

Popular models that run on Mac with 24 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.