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

Can I run NVIDIA-Nemotron-3-Nano-4B-BF16 on an RTX 3090?

Yes

Its published weights (7.95 GB) fit in the 21.6 GB an RTX 3090 can use, with room for an unknown amount of context.

Full analysis of NVIDIA-Nemotron-3-Nano-4B-BF16 Everything that runs on RTX 3090 Get the weights on Hugging Face
hf download nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16

Files and context on RTX 3090

Each file of NVIDIA-Nemotron-3-Nano-4B-BF16 (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 RTX 3090 Context (16-bit)Context (8-bit)
As published7.95 GBFitsnot knownnot known
4-bit (estimate)2.24 GBFitsnot knownnot known

NVIDIA-Nemotron-3-Nano-4B-BF16 on other hardware

Popular models that run on RTX 3090

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.