Community Benchmarks
Download oMLX →Real performance numbers from oMLX users around the world. Submit yours from the app (v0.2.6+).
Reset filters| # | Chip | RAM | Model | Quant | Ctx | PP tok/s | TG tok/s | Date ↓ |
|---|---|---|---|---|---|---|---|---|
| 161 | M3 Max (40c) | 64 GB | Qwen3.6-35B-A3B | 4bit | 64k | 908.2 | 62.2 | 26-07-24 |
| 162 | M3 Max (40c) | 64 GB | Qwen3-Coder-30B-A3B-Instr... | 4bit | 64k | 343.8 | 25.0 | 26-07-24 |
| 163 | M5 Max (40c) | 128 GB | GLM-4.7-Flash | 8bit | 64k | 433.1 | 35.6 | 26-07-24 |
| 164 | M2 Ultra (60c) | 64 GB | Qwen3.5-27B-Claude-4.6-Op... | 4bit | 64k | 152.8 | 12.6 | 26-07-24 |
| 165 | M5 Pro (20c) | 64 GB | qwen3.6-35b-a3b | 6bit | 64k | 1,163 | 48.6 | 26-07-24 |
| 166 | M5 Max (40c) | 128 GB | Laguna-S-2.1-oQ6e | 6bit | 64k | 659.8 | 37.0 | 26-07-24 |
| 167 | M5 Max (40c) | 128 GB | Laguna-S-2.1-oQ4e-fast | 4bit | 64k | 753.8 | 42.7 | 26-07-24 |
| 168 | M5 Max (40c) | 128 GB | Laguna-S-2.1-oQ4e-fast | 4bit | 64k | 763.4 | 42.2 | 26-07-24 |
| 169 | M4 Pro (20c) | 48 GB | Qwen3.6-35B-A3B-oQ6-mtp | 6bit | 64k | 333.3 | 28.9 | 26-07-24 |
| 170 | M2 Max (30c) | 64 GB | gemma-4-26B-A4B-it-qat | 8bit | 64k | 392.3 | 28.7 | 26-07-24 |