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 ↓ |
|---|---|---|---|---|---|---|---|---|
| 206871 | M5 Max (40c) | 128 GB | Qwen3.6-35B-A3B | 8bit | 1k | 2,071 | 95.4 | 26-04-17 |
| 206872 | M4 Max (40c) | 48 GB | Ternary-Bonsai-8B | 2bit | 64k | 295.5 | 17.4 | 26-04-17 |
| 206873 | M4 Max (40c) | 48 GB | Ternary-Bonsai-8B | 2bit | 16k | 691.0 | 69.0 | 26-04-17 |
| 206874 | M4 Max (40c) | 48 GB | Ternary-Bonsai-8B | 2bit | 32k | 531.4 | 48.5 | 26-04-17 |
| 206875 | M4 Max (40c) | 48 GB | Ternary-Bonsai-8B | 2bit | 4k | 918.7 | 106.9 | 26-04-17 |
| 206876 | M4 Max (40c) | 48 GB | Ternary-Bonsai-8B | 2bit | 8k | 830.8 | 87.9 | 26-04-17 |
| 206877 | M4 Max (40c) | 48 GB | Ternary-Bonsai-8B | 2bit | 1k | 897.7 | 128.1 | 26-04-17 |
| 206878 | M5 Max (40c) | 128 GB | Qwen3.6-35B-A3B-oQ4 | 4bit | 64k | 2,231 | 63.4 | 26-04-17 |
| 206879 | M5 Max (40c) | 128 GB | Qwen3.6-35B-A3B-oQ4 | 4bit | 128k | 1,254 | 18.2 | 26-04-17 |
| 206880 | M5 Max (40c) | 128 GB | Qwen3.6-35B-A3B-oQ4 | 4bit | 32k | 3,210 | 87.9 | 26-04-17 |