RX 6900 XT meets the minimum VRAM requirement for Q4 inference of petals-team/StableBeluga2. Review the quantization breakdown below to see how higher precision settings impact VRAM and throughput.
RX 6900 XT can run petals-team/StableBeluga2 with Q4 quantization. At approximately 78 tokens/second, you can expect Good speed - acceptable for interactive use.
You have 12GB headroom, which is sufficient for system overhead and smooth operation.
| Quantization | VRAM needed | VRAM available | Estimated speed | Verdict |
|---|---|---|---|---|
| Q4 | 4GB | 16GB | 77.61 tok/s | ✅ Fits comfortably |
| Q8 | 7GB | 16GB | 60.36 tok/s | ✅ Fits comfortably |
| FP16 | 15GB | 16GB | 28.89 tok/s | ⚠️ Tight fit |