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

Can I run gemma-3-27b-it-int4-awq on an NVIDIA B200?

Yes

Its published weights (18.5 GB) fit in the 162 GB an NVIDIA B200 can use, with room for ≈ 276K tokens of context.

Full analysis of gemma-3-27b-it-int4-awq Everything that runs on NVIDIA B200 Get the weights on Hugging Face
hf download gaunernst/gemma-3-27b-it-int4-awq

Files and context on NVIDIA B200

Each file of gemma-3-27b-it-int4-awq (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 B200 Context (16-bit)Context (8-bit)
As published18.5 GBFits276K552K
4-bit (estimate)15.4 GBFits282K564K

gemma-3-27b-it-int4-awq on other hardware

Popular models that run on NVIDIA B200

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.