Apple M2 Max meets the minimum VRAM requirement for Q4 inference of codellama/CodeLlama-34b-hf. Review the quantization breakdown below to see how higher precision settings impact VRAM and throughput.
Apple M2 Max can run codellama/CodeLlama-34b-hf with Q4 quantization. At approximately 20 tokens/second, you can expect Basic speed - best for non-interactive tasks.
You have 79GB headroom, which is sufficient for system overhead and smooth operation.
| Quantization | VRAM needed | VRAM available | Estimated speed | Verdict |
|---|---|---|---|---|
| Q4 | 17GB | 96GB | 19.93 tok/s | ✅ Fits comfortably |
| Q8 | 35GB | 96GB | 14.03 tok/s | ✅ Fits comfortably |
| FP16 | 70GB | 96GB | 6.54 tok/s | ✅ Fits comfortably |