Written by Jakub Rusinowski · Last updated April 22, 2026
Yes, but it is tight
Yes, but it is tight — Qwen 3.6 35B-A3B at Q4_K_M needs about 23.8 GB of the 24 GB usable on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM), leaving only ~0.2 GB before the runtime starts swapping. Expect ~169.9 tok/s (estimated), with room for about 8,192 tokens of context.
Confidence: low · Recommended quantization: Q4_K_M · Estimated speed: ~169.9 tok/s
| Usable memory for models | 24 GB |
| Memory bandwidth | 1008 GB/s |
| Form factor | Desktop |
| Operating system | Windows or Linux |
| Memory upgradeable | Yes |
| Quant | Memory needed | Fits 24 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 72.7 GB | ✗ No | — | — | 70 GB |
| Q8_0 | 39.9 GB | ✗ No | — | — | 37.2 GB |
| Q6_K | 31.4 GB | ✗ No | — | — | 28.7 GB |
| Q5_K_M | 27.5 GB | ✗ No | — | — | 24.8 GB |
| Q4_K_M | 23.8 GB | ✓ Yes | 8K | ~169.9 tok/s | 21.1 GB |
| Q3_K_M | 17.6 GB | ✓ Yes | 32K | ~193 tok/s | 14.9 GB |
| Q2_K | 14.2 GB | ✓ Yes | 32K | ~208.6 tok/s | 11.5 GB |
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|---|---|---|
| Qwen 3.6 35B-A3B | 23.8 GB | ✓ Fits | ~169.9 tok/s |
| Qwen 3.6 27B | 19.3 GB | ✓ Fits | ~39.1 tok/s |
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
Yes, but it is tight — Qwen 3.6 35B-A3B at Q4_K_M needs about 23.8 GB of the 24 GB usable on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM), leaving only ~0.2 GB before the runtime starts swapping. Expect ~169.9 tok/s (estimated), with room for about 8,192 tokens of context.
Q4_K_M — it needs about 23.8 GB of the 24 GB available, downloads as roughly 21.1 GB, and runs at an estimated 169.9 tokens/sec with up to 8K of context.
Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving