RX 9070 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.
RX 9070 can run openai/gpt-oss-20b with Q4 quantization. At approximately 48 tokens/second, you can expect Moderate speed - useful for batch processing.
You have 6GB headroom, which is sufficient for system overhead and smooth operation.
| Quantization | VRAM needed | VRAM available | Estimated speed | Verdict |
|---|---|---|---|---|
| Q4 | 10GB | 16GB | 48.00 tok/s | ✅ Fits comfortably |
| Q8 | 20GB | 16GB | 34.70 tok/s | ❌ Not recommended |
| FP16 | 41GB | 16GB | 19.86 tok/s | ❌ Not recommended |