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

Can I run Nanbeige4.3-33B on an NVIDIA B300?

Yes

Its published weights (66.2 GB) fit in the 259 GB an NVIDIA B300 can use, with room for ≈ 512K tokens of context.

Full analysis of Nanbeige4.3-33B Everything that runs on NVIDIA B300 Get the weights on Hugging Face
hf download Nanbeige/Nanbeige4.3-33B

Files and context on NVIDIA B300

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 NVIDIA B300 Context (16-bit)Context (8-bit)
As published66.2 GBFits512K512K
BF16bartowski/Nanbeige_Nanbeige4.3-33B-GGUF66.2 GBFits512K512K
Q8_0bartowski/Nanbeige_Nanbeige4.3-33B-GGUF35.2 GBFits512K512K
Q6_K_Lbartowski/Nanbeige_Nanbeige4.3-33B-GGUF30.4 GBFits512K512K
Q6_Kbartowski/Nanbeige_Nanbeige4.3-33B-GGUF29.0 GBFits512K512K
Q6_K_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF27.8 GBFits512K512K
Q5_K_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF25.1 GBFits512K512K
Q5_K_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF23.7 GBFits512K512K
Q4_K_Lbartowski/Nanbeige_Nanbeige4.3-33B-GGUF22.3 GBFits512K512K
Q4_1bartowski/Nanbeige_Nanbeige4.3-33B-GGUF21.2 GBFits512K512K
IQ4_NLbartowski/Nanbeige_Nanbeige4.3-33B-GGUF20.9 GBFits512K512K
Q4_K_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF20.8 GBFits512K512K
Q4_K_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF19.7 GBFits512K512K
Q4_0bartowski/Nanbeige_Nanbeige4.3-33B-GGUF19.3 GBFits512K512K
IQ4_XSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF18.6 GBFits512K512K
IQ3_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF17.8 GBFits512K512K
Q3_K_Lbartowski/Nanbeige_Nanbeige4.3-33B-GGUF17.3 GBFits512K512K
Q3_K_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF15.9 GBFits512K512K
IQ3_XSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF15.2 GBFits512K512K
Q3_K_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF15.2 GBFits512K512K
IQ3_XXSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF14.5 GBFits512K512K
Q2_Kbartowski/Nanbeige_Nanbeige4.3-33B-GGUF13.5 GBFits512K512K
IQ2_Mbartowski/Nanbeige_Nanbeige4.3-33B-GGUF12.2 GBFits512K512K
IQ2_Sbartowski/Nanbeige_Nanbeige4.3-33B-GGUF11.2 GBFits512K512K
IQ2_XSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF10.6 GBFits512K512K
IQ2_XXSbartowski/Nanbeige_Nanbeige4.3-33B-GGUF10.1 GBFits512K512K
Q8_0AtomicChat/Nanbeige4.3-33B-GGUF35.2 GBFits512K512K
Q6_KAtomicChat/Nanbeige4.3-33B-GGUF30.1 GBFits512K512K
Q5_KAtomicChat/Nanbeige4.3-33B-GGUF27.9 GBFits512K512K
Q5_K_MAtomicChat/Nanbeige4.3-33B-GGUF24.7 GBFits512K512K
Nanbeige4.3-33B-Q5_K_MAtomicChat/Nanbeige4.3-33B-GGUF23.5 GBFits512K512K
Nanbeige4.3-33B-AD-Q5_K_M-Q4_K_MAtomicChat/Nanbeige4.3-33B-GGUF22.6 GBFits512K512K
Q4_K_MAtomicChat/Nanbeige4.3-33B-GGUF21.2 GBFits512K512K
IQ4_XSAtomicChat/Nanbeige4.3-33B-GGUF20.3 GBFits512K512K
IQ3_SAtomicChat/Nanbeige4.3-33B-GGUF17.1 GBFits512K512K
IQ3_XXSAtomicChat/Nanbeige4.3-33B-GGUF14.9 GBFits512K512K
IQ2_SAtomicChat/Nanbeige4.3-33B-GGUF13.7 GBFits512K512K

Nanbeige4.3-33B on other hardware

Popular models that run on NVIDIA B300

Assumes 90% of the card's memory is usable 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.