NVIDIA L40S meets the minimum VRAM requirement for Q4 inference of Qwen/Qwen3-4B-Thinking-2507-FP8. Review the quantization breakdown below to see how higher precision settings impact VRAM and throughput.
NVIDIA L40S can run Qwen/Qwen3-4B-Thinking-2507-FP8 with Q4 quantization. At approximately 175 tokens/second, you can expect Excellent speed - conversational response times under 1 second.
You have 46GB headroom, which is sufficient for system overhead and smooth operation.
| Quantization | VRAM needed | VRAM available | Estimated speed | Verdict |
|---|---|---|---|---|
| Q4 | 2GB | 48GB | 174.51 tok/s | ✅ Fits comfortably |
| Q8 | 4GB | 48GB | 119.26 tok/s | ✅ Fits comfortably |
| FP16 | 9GB | 48GB | 58.97 tok/s | ✅ Fits comfortably |