Can I run Winnow-12B on an RTX 5070?

Yes

Its Winnow-12B-NVFP4 file (8.16 GB) fits in the 10.8 GB an RTX 5070 can use, with room for ≈ 15K tokens of context.

Full analysis of Winnow-12B Everything that runs on RTX 5070 Get the Winnow-12B-NVFP4 file on Hugging Face

ollama run hf.co/EldanRing/Winnow-12B

Files and context on RTX 5070

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.

File	Size	On RTX 5070	Context (16-bit)	Context (8-bit)
Winnow-12B-BF16	23.8 GB	Doesn't fit	—	—
Q8_0	12.7 GB	Doesn't fit	—	—
Winnow-12B-NVFP4	8.16 GB	Fits	15K	34K

Winnow-12B on other hardware

NVIDIA H100Yes
RTX 5090Yes, just
RTX 4090Yes
RTX 3090Yes
RTX 5060 Ti 16 GBYes, just
RTX 5070Yes
RTX 3060 12 GBYes
RTX 4060No
Mac with 32 GBYes
Mac with 64 GBYes
Mac with 128 GBYes

Popular models that run on RTX 5070

Qwen3-0.6BQwen · 752M
gpt2openai-community · 137M
Qwen3-Coder-30B-A3B-Instruct-GGUFunsloth · 30.5B
Qwen2.5-0.5B-InstructQwen · 494M
Qwen2.5-1.5B-InstructQwen · 1.54B
Qwen3-4BQwen · 4.02B
Qwen3.5-4BQwen · 4.66B
Ornith-1.5-9B-GGUFornith-ai · 9.20B

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.