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 ↓ |
|---|---|---|---|---|---|---|---|---|
| 205621 | M4 Max (40c) | 128 GB | Qwen3-Coder-Next | 8bit | 1k | 1,049 | 35.3 | 26-04-12 |
| 205622 | M4 Max (40c) | 128 GB | Qwen3-Coder-Next | 8bit | 4k | 1,321 | 65.6 | 26-04-12 |
| 205623 | M2 Ultra (60c) | 192 GB | MiniMax-M2.7-5bit | 5bit | 4k | 265.3 | 29.1 | 26-04-12 |
| 205624 | M2 Ultra (60c) | 192 GB | MiniMax-M2.7-5bit | 5bit | 1k | 227.8 | 35.3 | 26-04-12 |
| 205625 | M3 (10c) | 16 GB | gemma-4-e4b-it | 4bit | 4k | 281.1 | 12.0 | 26-04-12 |
| 205626 | M3 (10c) | 16 GB | gemma-4-e4b-it | 4bit | 1k | 233.8 | 12.9 | 26-04-12 |
| 205627 | M4 Max (40c) | 128 GB | Qwen3.5-35B-A3B | 4bit | 8k | 1,724 | 109.0 | 26-04-12 |
| 205628 | M4 Max (40c) | 128 GB | Qwen3.5-35B-A3B | 4bit | 1k | 1,393 | 120.8 | 26-04-12 |
| 205629 | M4 Max (40c) | 128 GB | Qwen3.5-35B-A3B | 4bit | 4k | 1,735 | 115.6 | 26-04-12 |
| 205630 | M4 Max (40c) | 128 GB | GLM-4.7-Flash | 8bit | 16k | 650.0 | 41.7 | 26-04-12 |