AMD Instinct MI250X meets the minimum VRAM requirement for Q4 inference of xgen-universe/Capybara. Review the quantization breakdown below to see how higher precision settings impact VRAM and throughput.
AMD Instinct MI250X can run xgen-universe/Capybara with Q4 quantization. At approximately 497 tokens/second, you can expect Excellent speed - conversational response times under 1 second.
You have 124GB headroom, which is sufficient for system overhead and smooth operation.
| Quantization | VRAM needed | VRAM available | Estimated speed | Verdict |
|---|---|---|---|---|
| Q4 | 4GB | 128GB | 497.22 tok/s | ✅ Fits comfortably |
| Q8 | 8GB | 128GB | 308.48 tok/s | ✅ Fits comfortably |
| FP16 | 16GB | 128GB | 174.19 tok/s | ✅ Fits comfortably |