Community Benchmarks
Download oMLX →Real performance numbers from oMLX users around the world. Submit yours from the app (v0.2.6+).
| # | Chip | RAM | Model | Quant | Ctx | PP tok/s | TG tok/s | Date ↓ |
|---|---|---|---|---|---|---|---|---|
| 205101 | M4 Max (40c) | 128 GB | NVIDIA-Nemotron-3-Super-1... | 4bit | 128k | 317.7 | 14.1 | 26-04-05 |
| 205102 | M4 Max (40c) | 128 GB | NVIDIA-Nemotron-3-Super-1... | 4bit | 64k | 368.5 | 22.8 | 26-04-05 |
| 205103 | M4 Max (40c) | 128 GB | NVIDIA-Nemotron-3-Super-1... | 4bit | 32k | 393.0 | 32.9 | 26-04-05 |
| 205104 | M4 Max (40c) | 128 GB | NVIDIA-Nemotron-3-Super-1... | 4bit | 16k | 395.2 | 38.8 | 26-04-05 |
| 205105 | M4 Max (40c) | 128 GB | NVIDIA-Nemotron-3-Super-1... | 4bit | 8k | 400.7 | 43.1 | 26-04-05 |
| 205106 | M4 Max (40c) | 128 GB | NVIDIA-Nemotron-3-Super-1... | 4bit | 4k | 405.7 | 47.5 | 26-04-05 |
| 205107 | M4 Max (40c) | 128 GB | NVIDIA-Nemotron-3-Super-1... | 4bit | 1k | 371.4 | 49.4 | 26-04-05 |
| 205108 | M1 Max (32c) | 32 GB | Qwen3.5-4B | 4bit | 4k | 274.5 | 44.8 | 26-04-05 |
| 205109 | M1 Max (32c) | 32 GB | Qwen3.5-4B | 4bit | 1k | 270.4 | 47.1 | 26-04-05 |
| 205110 | M1 Max (32c) | 64 GB | gemma-4-31b-it | 6bit | 16k | 69.6 | 5.8 | 26-04-05 |