Can RTX 4090 24GB run Qwen3-32B?

NVIDIA · 24 GB GDDR6X · 1008 GB/s bandwidth · Alibaba · 32.8B · model ctx up to 128K

Partly — Qwen3-32B at Q4 fits the RTX 4090 24GB (24 GB GDDR6X) at 4K context (≈ 22.32 GB) but not at the default 8K (≈ 25.6 GB needed). Use a shorter context or partial offload. Decode at Q4/8K: ≈ 49.1 tok/s (estimated, batch 1). FP16 needs ≈ 73.16 GB — does not fit on this card. Estimates come from memory-bandwidth math; rows tagged measured override estimates. Relative ranking is more reliable than absolute tok/s. Methodology.

Fit & speed by quant and context

Quant	Context	VRAM needed	Fits?	tok/s (decode)
Q4	4K	22.32 GB	Yes	19.6 est.
Q4	8K default	25.6 GB	No	49.1 est.
Q4	32K	45.28 GB	No	15.7 est.
FP16	4K	69.88 GB	No	1.8 est.
FP16	8K default	73.16 GB	No	1.7 est.
FP16	32K	92.84 GB	No	1.0 est.

Fits = weights + KV(ctx) + 1 GB overhead ≤ 95% of VRAM. Measured anchors are context-agnostic; the fit verdict is recomputed per context. A missing tok/s means the model is far beyond this card (offload-only territory).

VRAM breakdown at 8K context

Quant	Weights	KV cache	Overhead	Total needed	RTX 4090 24GB VRAM
Q4	18.04 GB	6.56 GB	1 GB	25.6 GB	24 GB
FP16	65.6 GB	6.56 GB	1 GB	73.16 GB	24 GB

Other GPUs that run Qwen3-32B

• B200 192GB SXM — ≈155.2 tok/s Q4 est.
• MI300X 192GB — ≈102.8 tok/s Q4 est.
• H200 141GB SXM5 — ≈93.1 tok/s Q4 est.
• H100 80GB SXM5 — ≈65.0 tok/s Q4 est.
• Gaudi 3 128GB — ≈63.6 tok/s Q4 est.
• A100 80GB SXM4 — ≈39.6 tok/s Q4 est.
• RTX 5090 32GB — ≈34.8 tok/s Q4 est.
• RTX 6000 Ada 48GB — ≈18.6 tok/s Q4 est.

All GPUs for Qwen3-32B →

Other models for the RTX 4090 24GB

• Qwen3-0.6B — ≈801.8 tok/s Q4 est.
• gemma-3-1b-it — ≈641.5 tok/s Q4 est.
• Llama-3.2-1B-Instruct — ≈534.5 tok/s Q4 est.
• Qwen3-1.7B — ≈320.7 tok/s Q4 est.
• gpt-oss-20b — ≈225.2 tok/s Q4 measured
• Qwen3-30B-A3B — ≈213.8 tok/s Q4 est.
• Qwen3-Coder-30B-A3B-Instruct — ≈213.8 tok/s Q4 est.
• GLM-4.7-Flash — ≈213.8 tok/s Q4 est.

All models on the RTX 4090 24GB →