Autor: Jakub Rusinowski · Ostatnia aktualizacja: 11 września 2026
Yes. Qwen 2.5 Coder 32B needs 19.3 GB at Q4_K_M and the NVIDIA GeForce RTX 3090 has 24 GB, leaving about 4.7 GB of headroom, at roughly 27 tok/s (estimated) with room for up to 8K of context.
See what else this hardware can run →
| Model | Qwen 2.5 Coder 32B |
| Parameters | 32B |
| Quantization | Q4_K_M |
| VRAM needed | 19.3 GB |
| NVIDIA GeForce RTX 3090 VRAM | 24 GB |
| Verdict | Runs |
| Max context | 8K |
| Estimated speed | ~27 tok/s |
| Quant | VRAM needed | Fits 24 GB? | Max context | Whole PC |
|---|---|---|---|---|
| F16 | 65.9 GB | ✗ No | — | — |
| Q8_0 | 35.9 GB | ✗ No | — | — |
| Q6_K | 28.1 GB | ✗ No | — | — |
| Q5_K_M | 24.6 GB | ✗ No | — | — |
| Q4_K_M | 21.2 GB | ✓ Yes | 8K | — |
| Q3_K_M | 15.5 GB | ✓ Yes | 32K | — |
| Q2_K | 12.4 GB | ✓ Yes | 32K | — |
Assumes a 4K-token context with an f16 KV cache. A longer window needs more; a quantized KV cache needs less. “Max context” is the largest window that still fits in 24 GB. Figures are estimates from parameter count, quantization and memory bandwidth — the analyzer lets you tune KV-cache quant and context.
Context costs VRAM too. At Q4_K_M on the NVIDIA GeForce RTX 3090: 8K 22.3 GB · 32K 28.7 GB ✗ · 128K 54.5 GB ✗. Past 32K it no longer fits 24 GB — a q8_0 KV cache buys roughly half of that back, which the calculator will price for you.
Nothing to change — run it at Q4_K_M. If you want more capability, check whether the next size up also fits.
No compatible GPU? Qwen 2.5 Coder 32B on 32 GB of system RAM, CPU only: runs comfortably.
Running Qwen 2.5 Coder 32B on this card:
At 2 hrs/day, buying (~$1,499) beats renting at $0.34/hr after about 6.1 years.
Affiliate links — we may earn a commission if you sign up, at no extra cost to you.
Cloud rates verified 2026-07 — estimates, and marketplace prices vary. Buying price is GPU MSRP only, not a full PC.
All Qwen 2.5 Family sizes on the NVIDIA GeForce RTX 3090 | Qwen 2.5 Coder 32B model page | NVIDIA GeForce RTX 3090 GPU page | Check your hardware