AMD Instinct MI300X meets the minimum VRAM requirement for Q4 inference of ibm-granite/granite-3.3-8b-instruct. Review the quantization breakdown below to see how higher precision settings impact VRAM and throughput.
AMD Instinct MI300X can run ibm-granite/granite-3.3-8b-instruct with Q4 quantization. At approximately 839 tokens/second, you can expect Excellent speed - conversational response times under 1 second.
You have 188GB headroom, which is sufficient for system overhead and smooth operation.
| Quantization | VRAM needed | VRAM available | Estimated speed | Verdict |
|---|---|---|---|---|
| Q4 | 4GB | 192GB | 838.91 tok/s | ✅ Fits comfortably |
| Q8 | 9GB | 192GB | 532.25 tok/s | ✅ Fits comfortably |
| FP16 | 17GB | 192GB | 311.42 tok/s | ✅ Fits comfortably |