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

Can I run Winnow-12B on an RTX 3060 12 GB?

Yes

Its Winnow-12B-NVFP4 file (8.16 GB) fits in the 10.8 GB an RTX 3060 12 GB can use, with room for ≈ 15K tokens of context.

Full analysis of Winnow-12B Everything that runs on RTX 3060 12 GB Get the Winnow-12B-NVFP4 file on Hugging Face
ollama run hf.co/EldanRing/Winnow-12B

Files and context on RTX 3060 12 GB

Each file of Winnow-12B (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 3060 12 GB Context (16-bit)Context (8-bit)
Winnow-12B-BF1623.8 GBDoesn't fit——
Q8_012.7 GBDoesn't fit——
Winnow-12B-NVFP48.16 GBFits15K34K

Winnow-12B on other hardware

Popular models that run on RTX 3060 12 GB

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.