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 ↓ |
|---|---|---|---|---|---|---|---|---|
| 205011 | M4 (10c) | 16 GB | Qwen3.5-9B | 4bit | 4k | 222.5 | 20.5 | 26-04-05 |
| 205012 | M1 Max (24c) | 32 GB | gemma-4-26b-a4b-it | 4bit | 4k | 364.6 | 42.7 | 26-04-05 |
| 205013 | M1 Max (24c) | 32 GB | gemma-4-26b-a4b-it | 4bit | 1k | 322.3 | 45.9 | 26-04-05 |
| 205014 | M5 Max (40c) | 128 GB | gpt-oss-120b-MXFP4-Q8 | 4bit | 128k | 521.2 | 14.0 | 26-04-05 |
| 205015 | M5 Max (40c) | 128 GB | gpt-oss-120b-MXFP4-Q8 | 4bit | 195k | 404.2 | 6.3 | 26-04-05 |
| 205016 | M5 Max (40c) | 128 GB | gpt-oss-120b-MXFP4-Q8 | 4bit | 64k | 646.7 | 22.5 | 26-04-05 |
| 205017 | M5 Max (40c) | 128 GB | gpt-oss-120b-MXFP4-Q8 | 4bit | 32k | 1,213 | 41.4 | 26-04-05 |
| 205018 | M5 Max (40c) | 128 GB | gpt-oss-120b-MXFP4-Q8 | 4bit | 8k | 1,658 | 70.6 | 26-04-05 |
| 205019 | M5 Max (40c) | 128 GB | gpt-oss-120b-MXFP4-Q8 | 4bit | 16k | 1,538 | 58.5 | 26-04-05 |
| 205020 | M5 Max (40c) | 128 GB | gpt-oss-120b-MXFP4-Q8 | 4bit | 1k | 1,023 | 79.7 | 26-04-05 |