RTX 5090 meets the minimum VRAM requirement for Q4 inference of openai/gpt-oss-20b. Review the quantization breakdown below to see how higher precision settings impact VRAM and throughput.
RTX 5090 can run openai/gpt-oss-20b with Q4 quantization. At approximately 157 tokens/second, you can expect Excellent speed - conversational response times under 1 second.
You have 22GB headroom, which is sufficient for system overhead and smooth operation.
| Quantization | VRAM needed | VRAM available | Estimated speed | Verdict |
|---|---|---|---|---|
| Q4 | 10GB | 32GB | 156.71 tok/s | ✅ Fits comfortably |
| Q8 | 20GB | 32GB | 126.70 tok/s | ✅ Fits comfortably |
| FP16 | 41GB | 32GB | 58.13 tok/s | ❌ Not recommended |