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 ↓ |
|---|---|---|---|---|---|---|---|---|
| 71331 | M4 (10c) | 16 GB | Qwen3.5-4B | 4bit | 1k | 384.9 | 38.4 | 26-05-15 |
| 71332 | M4 (10c) | 16 GB | Qwen3.5-4B | 4bit | 4k | 397.2 | 36.8 | 26-05-15 |
| 71333 | M2 Ultra (60c) | 192 GB | gemma-4-31B-it | 8bit | 4k | 146.6 | 16.2 | 26-05-15 |
| 71334 | M2 Ultra (60c) | 192 GB | gemma-4-31B-it | 8bit | 1k | 145.5 | 16.8 | 26-05-15 |
| 71335 | M2 Max (30c) | 64 GB | Qwen3.6-35B-A3B | 4bit | 16k | 604.0 | 64.2 | 26-05-15 |
| 71336 | M2 Max (30c) | 64 GB | Qwen3.6-35B-A3B | 4bit | 4k | 628.9 | 75.0 | 26-05-15 |
| 71337 | M2 Max (38c) | 96 GB | Qwen3.6-35B-A3B-oQ6-fp16-... | 6bit | 8k | 841.8 | 56.6 | 26-05-15 |
| 71338 | M2 Max (38c) | 96 GB | Qwen3.6-35B-A3B-oQ6-fp16-... | 6bit | 8k | 842.7 | 56.2 | 26-05-15 |
| 71339 | M2 Max (30c) | 64 GB | Qwen3.6-35B-A3B-oQ8 | 8bit | 16k | 1,032 | 60.4 | 26-05-15 |
| 71340 | M2 Max (30c) | 64 GB | Qwen3.6-35B-A3B-oQ8 | 8bit | 4k | 1,106 | 66.6 | 26-05-15 |