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

Can I run surya-ocr-2 on an RTX 4090?

Yes

Its published weights (1.37 GB) fit in the 21.6 GB an RTX 4090 can use, with room for ≈ 256K tokens of context.

Full analysis of surya-ocr-2 Everything that runs on RTX 4090 Get the weights on Hugging Face
hf download datalab-to/surya-ocr-2

Files and context on RTX 4090

Each file of surya-ocr-2 (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 4090 Context (16-bit)Context (8-bit)
As published1.37 GBFits256K256K
surya-2datalab-to/surya-ocr-2-gguf1.27 GBFits256K256K

surya-ocr-2 on other hardware

Popular models that run on RTX 4090

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.