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

Can I run BigBang-v1 on an NVIDIA H100?

Yes, with a GGUF version

The published weights are too big, but the BF16 file from bartowski/endless-frontier_BigBang-v1-GGUF (71.1 GB) fits, with room for ≈ 41K tokens of context.

Full analysis of BigBang-v1 Everything that runs on NVIDIA H100 Get the BF16 file on Hugging Face
llama-server -hf bartowski/endless-frontier_BigBang-v1-GGUF:BF16

Files and context on NVIDIA H100

Each file of BigBang-v1 (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)
As published71.9 GBDoesn't fit2K3K
BF16bartowski/endless-frontier_BigBang-v1-GGUF71.1 GBTight41K83K
Q8_0bartowski/endless-frontier_BigBang-v1-GGUF37.8 GBFits256K256K
Q6_K_Lbartowski/endless-frontier_BigBang-v1-GGUF30.8 GBFits256K256K
Q6_Kbartowski/endless-frontier_BigBang-v1-GGUF30.5 GBFits256K256K
Q5_K_Lbartowski/endless-frontier_BigBang-v1-GGUF25.8 GBFits256K256K
Q5_K_Mbartowski/endless-frontier_BigBang-v1-GGUF25.5 GBFits256K256K
Q5_K_Sbartowski/endless-frontier_BigBang-v1-GGUF24.6 GBFits256K256K
Q4_1bartowski/endless-frontier_BigBang-v1-GGUF22.4 GBFits256K256K
Q4_K_Lbartowski/endless-frontier_BigBang-v1-GGUF22.2 GBFits256K256K
Q4_K_Mbartowski/endless-frontier_BigBang-v1-GGUF21.9 GBFits256K256K
Q4_K_Sbartowski/endless-frontier_BigBang-v1-GGUF21.1 GBFits256K256K
Q4_0bartowski/endless-frontier_BigBang-v1-GGUF20.4 GBFits256K256K
IQ4_NLbartowski/endless-frontier_BigBang-v1-GGUF20.3 GBFits256K256K
IQ4_XSbartowski/endless-frontier_BigBang-v1-GGUF19.3 GBFits256K256K
Q3_K_XLbartowski/endless-frontier_BigBang-v1-GGUF17.8 GBFits256K256K
IQ3_Mbartowski/endless-frontier_BigBang-v1-GGUF17.4 GBFits256K256K
Q3_K_Lbartowski/endless-frontier_BigBang-v1-GGUF17.4 GBFits256K256K
Q3_K_Mbartowski/endless-frontier_BigBang-v1-GGUF16.7 GBFits256K256K
IQ3_XSbartowski/endless-frontier_BigBang-v1-GGUF16.7 GBFits256K256K
Q3_K_Sbartowski/endless-frontier_BigBang-v1-GGUF16.0 GBFits256K256K
IQ3_XXSbartowski/endless-frontier_BigBang-v1-GGUF15.3 GBFits256K256K
Q2_K_Lbartowski/endless-frontier_BigBang-v1-GGUF13.6 GBFits256K256K
Q2_Kbartowski/endless-frontier_BigBang-v1-GGUF13.1 GBFits256K256K
IQ2_Mbartowski/endless-frontier_BigBang-v1-GGUF12.5 GBFits256K256K
IQ2_Sbartowski/endless-frontier_BigBang-v1-GGUF11.5 GBFits256K256K
IQ2_XSbartowski/endless-frontier_BigBang-v1-GGUF11.3 GBFits256K256K
IQ2_XXSbartowski/endless-frontier_BigBang-v1-GGUF10.3 GBFits256K256K

BigBang-v1 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.