Apple M2 Max meets the minimum VRAM requirement for Q4 inference of kaitchup/Phi-3-mini-4k-instruct-gptq-4bit. Review the quantization breakdown below to see how higher precision settings impact VRAM and throughput.
Apple M2 Max can run kaitchup/Phi-3-mini-4k-instruct-gptq-4bit with Q4 quantization. At approximately 52 tokens/second, you can expect Good speed - acceptable for interactive use.
You have 94GB headroom, which is sufficient for system overhead and smooth operation.
| Quantization | VRAM needed | VRAM available | Estimated speed | Verdict |
|---|---|---|---|---|
| Q4 | 2GB | 96GB | 52.10 tok/s | ✅ Fits comfortably |
| Q8 | 4GB | 96GB | 39.75 tok/s | ✅ Fits comfortably |
| FP16 | 9GB | 96GB | 21.53 tok/s | ✅ Fits comfortably |