Consumer NVIDIA
Figures assume this GPU plus 32 GB of system RAM (a typical desktop pairing). "What runs on it" is judged at a 8,192-token context. Speeds are estimates, not measurements.
| Model | Sweet-spot quant | Est. speed | Community |
|---|---|---|---|
| Mixtral-8x7B-Instruct-v0.1 mistralai | Q4_K_M fits on GPU | 65–86 tok/s (estimate) | no community data |
| Qwen2.5-7B-Instruct Qwen | Q4_K_M fits on GPU | 16–21 tok/s (estimate) | no community data |
| Qwen3-8B Qwen | Q4_K_M fits on GPU | 28–37 tok/s (estimate) | no community data |
| Qwen3-30B-A3B-Instruct-2507 Qwen | Q4_K_M CPU offload@if ($run['n_gpu_layers'] !== null) · 36 layers on GPU@endif | 34–46 tok/s (estimate) | no community data |
| Qwen3-32B Qwen | Q4_K_M CPU offload@if ($run['n_gpu_layers'] !== null) · 41 layers on GPU@endif | 4–5 tok/s (estimate) | no community data |
"Est. speed" is a modelled range labelled estimate (D8). "Community" shows the median of approved user-submitted reports on this GPU class only where enough exist — never an estimate.