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

Can I run occamy-1.0 on an NVIDIA H200?

Yes

Its published weights (70.2 GB) fit in the 127 GB an NVIDIA H200 can use, with room for ≈ 256K tokens of context.

Full analysis of occamy-1.0 Everything that runs on NVIDIA H200 Get the weights on Hugging Face
hf download Accio-Lab/occamy-1.0

Files and context on NVIDIA H200

Each file of occamy-1.0 (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 H200 Context (16-bit)Context (8-bit)
As published70.2 GBFits256K256K
Q8_0Accio-Lab/occamy-1.0-GGUF36.9 GBFits256K256K
Q6_KAccio-Lab/occamy-1.0-GGUF28.5 GBFits256K256K
Q5_K_MAccio-Lab/occamy-1.0-GGUF24.7 GBFits256K256K
Q5_0Accio-Lab/occamy-1.0-GGUF24.0 GBFits256K256K
Q5_K_SAccio-Lab/occamy-1.0-GGUF24.0 GBFits256K256K
Q4_K_MAccio-Lab/occamy-1.0-GGUF21.2 GBFits256K256K
Q4_K_SAccio-Lab/occamy-1.0-GGUF19.9 GBFits256K256K
IQ4_NLAccio-Lab/occamy-1.0-GGUF19.8 GBFits256K256K
Q4_0Accio-Lab/occamy-1.0-GGUF19.7 GBFits256K256K
IQ4_XSAccio-Lab/occamy-1.0-GGUF18.7 GBFits256K256K
Q3_K_LAccio-Lab/occamy-1.0-GGUF18.1 GBFits256K256K
Q3_K_MAccio-Lab/occamy-1.0-GGUF16.8 GBFits256K256K
IQ3_MAccio-Lab/occamy-1.0-GGUF15.4 GBFits256K256K
Q3_K_SAccio-Lab/occamy-1.0-GGUF15.2 GBFits256K256K
IQ3_XSAccio-Lab/occamy-1.0-GGUF14.5 GBFits256K256K
Q2_KAccio-Lab/occamy-1.0-GGUF12.9 GBFits256K256K
IQ2_MAccio-Lab/occamy-1.0-GGUF11.7 GBFits256K256K

occamy-1.0 on other hardware

Popular models that run on NVIDIA H200

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.