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

Can I run gemma-4-12b-it-GGUF on an NVIDIA H100?

Yes

Its gemma-4-12b-it-BF16 file (23.8 GB) fits in the 72.0 GB an NVIDIA H100 can use, with room for ≈ 256K tokens of context.

Full analysis of gemma-4-12b-it-GGUF Everything that runs on NVIDIA H100 Get the gemma-4-12b-it-BF16 file on Hugging Face
ollama run hf.co/unsloth/gemma-4-12b-it-GGUF

Files and context on NVIDIA H100

Each file of gemma-4-12b-it-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)
gemma-4-12b-it-BF1623.8 GBFits256K256K
UD-Q8_K_XL13.6 GBFits256K256K
gemma-4-12b-it-Q8_012.7 GBFits256K256K
UD-Q6_K_XL10.7 GBFits256K256K
Q6_K9.79 GBFits256K256K
UD-Q5_K_XL8.61 GBFits256K256K
Q5_K_M8.41 GBFits256K256K
Q5_K_S8.20 GBFits256K256K
Q4_17.40 GBFits256K256K
UD-Q4_K_XL7.37 GBFits256K256K
Q4_K_M7.12 GBFits256K256K
Q4_K_S6.76 GBFits256K256K
Q4_06.74 GBFits256K256K
IQ4_NL6.72 GBFits256K256K
IQ4_XS6.38 GBFits256K256K
UD-Q3_K_XL6.02 GBFits256K256K
Q3_K_M5.69 GBFits256K256K
Q3_K_S5.14 GBFits256K256K
UD-Q2_K_XL4.66 GBFits256K256K
UD-IQ3_XXS4.64 GBFits256K256K
UD-IQ2_M4.21 GBFits256K256K

gemma-4-12b-it-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.