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

Can I run Nanbeige4.3-33B on a Mac with 48 GB?

Yes, with a GGUF version

The published weights are too big, but the Q6_K_L file from bartowski/Nanbeige_Nanbeige4.3-33B-GGUF (30.4 GB) fits, with room for ≈ 26K tokens of context.

Full analysis of Nanbeige4.3-33B Everything that runs on Mac with 48 GB Get the Q6_K_L file on Hugging Face
ollama run hf.co/bartowski/Nanbeige_Nanbeige4.3-33B-GGUF:Q6_K_L

Files and context on Mac with 48 GB

Each file of Nanbeige4.3-33B (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 48 GB Context (16-bit)Context (8-bit)
As published66.2 GBDoesn't fit——
BF16bartowski/Nanbeige_Nanbeige4.3-33B-GGUF66.2 GBDoesn't fit——
Q8_0bartowski/Nanbeige_Nanbeige4.3-33B-GGUF35.2 GBDoesn't fit4K7K
Q6_K_Lbartowski/Nanbeige_Nanbeige4.3-33B-GGUF30.4 GBTight26K51K
Q6_Kbartowski/Nanbeige_Nanbeige4.3-33B-GGUF29.0 GBTight32K64K
Q6_K_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF27.8 GBFits38K76K
Q5_K_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF25.1 GBFits50K100K
Q5_K_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF23.7 GBFits57K113K
Q4_K_Lbartowski/Nanbeige_Nanbeige4.3-33B-GGUF22.3 GBFits63K125K
Q4_1bartowski/Nanbeige_Nanbeige4.3-33B-GGUF21.2 GBFits68K136K
IQ4_NLbartowski/Nanbeige_Nanbeige4.3-33B-GGUF20.9 GBFits69K139K
Q4_K_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF20.8 GBFits70K139K
Q4_K_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF19.7 GBFits75K149K
Q4_0bartowski/Nanbeige_Nanbeige4.3-33B-GGUF19.3 GBFits76K153K
IQ4_XSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF18.6 GBFits80K159K
IQ3_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF17.8 GBFits83K167K
Q3_K_Lbartowski/Nanbeige_Nanbeige4.3-33B-GGUF17.3 GBFits86K172K
Q3_K_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF15.9 GBFits92K184K
IQ3_XSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF15.2 GBFits95K191K
Q3_K_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF15.2 GBFits95K191K
IQ3_XXSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF14.5 GBFits99K197K
Q2_Kbartowski/Nanbeige_Nanbeige4.3-33B-GGUF13.5 GBFits103K206K
IQ2_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF12.2 GBFits109K218K
IQ2_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF11.2 GBFits114K228K
IQ2_XSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF10.6 GBFits116K233K
IQ2_XXSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF10.1 GBFits119K238K
Q8_0AtomicChat/Nanbeige4.3-33B-GGUF35.2 GBDoesn't fit4K7K
Q6_KAtomicChat/Nanbeige4.3-33B-GGUF30.1 GBTight27K54K
Q5_KAtomicChat/Nanbeige4.3-33B-GGUF27.9 GBFits37K74K
Q5_K_MAtomicChat/Nanbeige4.3-33B-GGUF24.7 GBFits52K104K
Nanbeige4.3-33B-Q5_K_MAtomicChat/Nanbeige4.3-33B-GGUF23.5 GBFits57K115K
Nanbeige4.3-33B-AD-Q5_K_M-Q4_K_MAtomicChat/Nanbeige4.3-33B-GGUF22.6 GBFits61K123K
Q4_K_MAtomicChat/Nanbeige4.3-33B-GGUF21.2 GBFits68K136K
IQ4_XSAtomicChat/Nanbeige4.3-33B-GGUF20.3 GBFits72K144K
IQ3_SAtomicChat/Nanbeige4.3-33B-GGUF17.1 GBFits87K174K
IQ3_XXSAtomicChat/Nanbeige4.3-33B-GGUF14.9 GBFits97K193K
IQ2_SAtomicChat/Nanbeige4.3-33B-GGUF13.7 GBFits102K204K

Nanbeige4.3-33B on other hardware

Popular models that run on Mac with 48 GB

Assumes 75% 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.