Qwen3 32B at Q4_K_M needs about 22.3 GB, more than the 16 GB of the RTX 5080: part of the model runs from system RAM, which slows generation. At Q3_K_M it does not fit in VRAM.
Q4_K_M · 8K context
The model needs about 22.3 GB but only 15.5 GB of VRAM is usable: 6.8 GB must be loaded into system RAM. It runs, 70% on the GPU, but more slowly.
~22.3 GB (21.6 GB–23.1 GB) needed · estimate
6.8 GB must be offloaded to system RAM
EstimateData updated 24 days agoSource: Run LLM Locally engine, Hugging Face config.json + model card
What makes this estimate less reliable
| Qwen3 32B Q8_08K~37 GB · 0.9–1.8 tok/s estimate | Not recommended |
|---|---|
| Qwen3 32B Q6_K8K~29.2 GB · 1.4–3 tok/s estimate | Possible but slow |
| Qwen3 32B Q5_K_M8K~25.7 GB · 1.9–4 tok/s estimate | Possible but slow |
| Qwen3 32B Q4_K_M8K~22.3 GB · 3–5.5 tok/s estimate | CPU/GPU offload |
| Qwen3 32B IQ4_XS8K~20.3 GB · 4–7.5 tok/s estimate | CPU/GPU offload |
| Qwen3 32B Q3_K_M8K~18.6 GB · 6–11 tok/s estimate | CPU/GPU offload |
Prices and links follow the market you choose.
AMD
CompatibleRuns Qwen3 32B Q4_K_M entirely in memory
Apple Silicon
CompatibleRuns Qwen3 32B Q4_K_M entirely in memory
Unified memory PC
CompatibleRuns Qwen3 32B Q4_K_M entirely in memory
No current offer imported for this market: the link opens a marketplace search.
