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 ↓ |
|---|---|---|---|---|---|---|---|---|
| 7411 | M4 Pro (20c) | 48 GB | Qwen3.5-9B | 4bit | 1k | 445.0 | 48.9 | 26-03-09 |
| 7412 | M4 Pro (20c) | 48 GB | Qwen3.5-9B | 4bit | 4k | 447.3 | 47.6 | 26-03-09 |
| 7413 | M4 Pro (20c) | 48 GB | gemma-3-4b-it-qat | 4bit | 4k | 889.9 | 80.2 | 26-03-09 |
| 7414 | M4 Pro (20c) | 48 GB | gemma-3-4b-it-qat | 4bit | 1k | 901.8 | 34.9 | 26-03-09 |
| 7415 | M1 Pro (16c) | 32 GB | Qwen3.5-4B | 4bit | 4k | 292.5 | 49.3 | 26-03-09 |
| 7416 | M1 Pro (16c) | 32 GB | Qwen3.5-4B | 4bit | 1k | 292.6 | 53.8 | 26-03-09 |
| 7417 | M1 Pro (16c) | 32 GB | Qwen3.5-9B | 4bit | 4k | 159.3 | 31.5 | 26-03-09 |
| 7418 | M1 Pro (16c) | 32 GB | Qwen3.5-9B | 4bit | 1k | 158.6 | 32.3 | 26-03-09 |
| 7419 | M2 Ultra (76c) | 128 GB | Qwen3.5-9B | 8bit | 32k | 742.0 | 46.4 | 26-03-09 |
| 7420 | M2 Ultra (76c) | 128 GB | Qwen3.5-9B | 8bit | 64k | 668.6 | 33.9 | 26-03-09 |