Can I run Qwen3-VL-32B-Thinking on an RTX 4090?

Yes, with a GGUF version

The published weights are too big, but the Q4_K_M file from Qwen/Qwen3-VL-32B-Thinking-GGUF (19.8 GB) fits, with room for ≈ 7K tokens of context.

Full analysis of Qwen3-VL-32B-Thinking Everything that runs on RTX 4090 Get the Q4_K_M file on Hugging Face

ollama run hf.co/Qwen/Qwen3-VL-32B-Thinking-GGUF:Q4_K_M

Files and context on RTX 4090

Each file of Qwen3-VL-32B-Thinking (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 4090	Context (16-bit)	Context (8-bit)
As published	66.7 GB	Doesn't fit	—	—
F16Qwen/Qwen3-VL-32B-Thinking-GGUF	65.5 GB	Doesn't fit	—	—
Q8_0Qwen/Qwen3-VL-32B-Thinking-GGUF	34.8 GB	Doesn't fit	—	—
Q4_K_MQwen/Qwen3-VL-32B-Thinking-GGUF	19.8 GB	Tight	7K	14K

Qwen3-VL-32B-Thinking 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.