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

Can I run Qwen3-Coder-30B-A3B-Instruct-GGUF on an NVIDIA H100?

Yes, just

Its BF16 file (61.1 GB) fits in the 72.0 GB an NVIDIA H100 can use, with room for ≈ 108K tokens of context.

Full analysis of Qwen3-Coder-30B-A3B-Instruct-GGUF Everything that runs on NVIDIA H100 Get the BF16 file on Hugging Face
llama-server -hf unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF:BF16

Files and context on NVIDIA H100

Each file of Qwen3-Coder-30B-A3B-Instruct-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 NVIDIA H100 Context (16-bit)Context (8-bit)
BF1661.1 GBTight108K217K
UD-Q8_K_XL36.0 GBFits256K256K
Q8_032.5 GBFits256K256K
UD-Q6_K_XL26.3 GBFits256K256K
Q6_K25.1 GBFits256K256K
UD-Q5_K_XL21.7 GBFits256K256K
Q5_K_M21.7 GBFits256K256K
Q5_K_S21.1 GBFits256K256K
Q4_119.2 GBFits256K256K
Q4_K_M18.6 GBFits256K256K
UD-Q4_K_XL17.7 GBFits256K256K
Q4_K_S17.5 GBFits256K256K
Q4_017.4 GBFits256K256K
IQ4_NL17.3 GBFits256K256K
IQ4_XS16.4 GBFits256K256K
Q3_K_M14.7 GBFits256K256K
UD-Q3_K_XL13.8 GBFits256K256K
Q3_K_S13.3 GBFits256K256K
UD-IQ3_XXS12.8 GBFits256K256K
UD-Q2_K_XL11.8 GBFits256K256K
Q2_K_L11.3 GBFits256K256K
Q2_K11.3 GBFits256K256K
UD-IQ2_M10.8 GBFits256K256K
UD-IQ2_XXS10.3 GBFits256K256K
UD-IQ1_M9.63 GBFits256K256K
UD-IQ1_S8.91 GBFits256K256K
UD-TQ1_08.01 GBFits256K256K

Qwen3-Coder-30B-A3B-Instruct-GGUF on other hardware

Popular models that run on NVIDIA H100

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.