Qwen2.5 Coder 32B Instruct at Q4_K_M needs about 22.4 GB, more than the 16 GB of the RTX 4060 Ti 16 GB: 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 22.4 GB but only 15.5 GB of VRAM is usable: 6.9 GB must be loaded into system RAM. It runs, 69% on the GPU, but more slowly.
Estimated speed is at most 4.5 tokens/s: usable, but slow.
~22.4 GB (21.6 GB–23.2 GB) needed · estimate
6.9 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
| Qwen2.5 Coder 32B Instruct Q8_08K~37 GB · 0.9–1.7 tok/s estimate | Not recommended |
|---|---|
| Qwen2.5 Coder 32B Instruct Q6_K8K~29.3 GB · 1.3–2.5 tok/s estimate | Possible but slow |
| Qwen2.5 Coder 32B Instruct Q5_K_M8K~25.7 GB · 1.7–3.5 tok/s estimate | Possible but slow |
| Qwen2.5 Coder 32B Instruct Q4_K_M8K~22.4 GB · 2.5–4.5 tok/s estimate | Possible but slow |
| Qwen2.5 Coder 32B Instruct IQ4_XS8K~20.3 GB · 3.5–6 tok/s estimate | CPU/GPU offload |
| Qwen2.5 Coder 32B Instruct Q3_K_M8K~18.5 GB · 4.5–8 tok/s estimate | CPU/GPU offload |
Prices and links follow the market you choose.
AMD
CompatibleRuns Qwen2.5 Coder 32B Instruct Q4_K_M entirely in memory
Apple Silicon
CompatibleRuns Qwen2.5 Coder 32B Instruct Q4_K_M entirely in memory
Unified memory PC
CompatibleRuns Qwen2.5 Coder 32B Instruct Q4_K_M entirely in memory
No current offer imported for this market: the link opens a marketplace search.
