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

Can I run Nanbeige4.3-33B on a Mac with 96 GB?

Yes, just

Its published weights (66.2 GB) fit in the 72.0 GB a Mac with 96 GB can use, with room for ≈ 26K tokens of context.

Full analysis of Nanbeige4.3-33B Everything that runs on Mac with 96 GB Get the weights on Hugging Face
hf download Nanbeige/Nanbeige4.3-33B

Files and context on Mac with 96 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 96 GB Context (16-bit)Context (8-bit)
As published66.2 GBTight26K53K
BF16bartowski/Nanbeige_Nanbeige4.3-33B-GGUF66.2 GBTight26K53K
Q8_0bartowski/Nanbeige_Nanbeige4.3-33B-GGUF35.2 GBFits169K337K
Q6_K_Lbartowski/Nanbeige_Nanbeige4.3-33B-GGUF30.4 GBFits191K381K
Q6_Kbartowski/Nanbeige_Nanbeige4.3-33B-GGUF29.0 GBFits197K394K
Q6_K_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF27.8 GBFits203K406K
Q5_K_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF25.1 GBFits215K430K
Q5_K_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF23.7 GBFits222K443K
Q4_K_Lbartowski/Nanbeige_Nanbeige4.3-33B-GGUF22.3 GBFits228K455K
Q4_1bartowski/Nanbeige_Nanbeige4.3-33B-GGUF21.2 GBFits233K466K
IQ4_NLbartowski/Nanbeige_Nanbeige4.3-33B-GGUF20.9 GBFits234K469K
Q4_K_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF20.8 GBFits235K470K
Q4_K_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF19.7 GBFits240K479K
Q4_0bartowski/Nanbeige_Nanbeige4.3-33B-GGUF19.3 GBFits241K483K
IQ4_XSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF18.6 GBFits245K489K
IQ3_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF17.8 GBFits248K497K
Q3_K_Lbartowski/Nanbeige_Nanbeige4.3-33B-GGUF17.3 GBFits251K502K
Q3_K_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF15.9 GBFits257K512K
IQ3_XSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF15.2 GBFits260K512K
Q3_K_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF15.2 GBFits260K512K
IQ3_XXSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF14.5 GBFits264K512K
Q2_Kbartowski/Nanbeige_Nanbeige4.3-33B-GGUF13.5 GBFits268K512K
IQ2_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF12.2 GBFits274K512K
IQ2_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF11.2 GBFits279K512K
IQ2_XSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF10.6 GBFits281K512K
IQ2_XXSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF10.1 GBFits284K512K
Q8_0AtomicChat/Nanbeige4.3-33B-GGUF35.2 GBFits169K337K
Q6_KAtomicChat/Nanbeige4.3-33B-GGUF30.1 GBFits192K384K
Q5_KAtomicChat/Nanbeige4.3-33B-GGUF27.9 GBFits202K404K
Q5_K_MAtomicChat/Nanbeige4.3-33B-GGUF24.7 GBFits217K434K
Nanbeige4.3-33B-Q5_K_MAtomicChat/Nanbeige4.3-33B-GGUF23.5 GBFits222K445K
Nanbeige4.3-33B-AD-Q5_K_M-Q4_K_MAtomicChat/Nanbeige4.3-33B-GGUF22.6 GBFits226K453K
Q4_K_MAtomicChat/Nanbeige4.3-33B-GGUF21.2 GBFits233K466K
IQ4_XSAtomicChat/Nanbeige4.3-33B-GGUF20.3 GBFits237K474K
IQ3_SAtomicChat/Nanbeige4.3-33B-GGUF17.1 GBFits252K504K
IQ3_XXSAtomicChat/Nanbeige4.3-33B-GGUF14.9 GBFits262K512K
IQ2_SAtomicChat/Nanbeige4.3-33B-GGUF13.7 GBFits267K512K

Nanbeige4.3-33B on other hardware

Popular models that run on Mac with 96 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.