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 ↓ |
|---|---|---|---|---|---|---|---|---|
| 42611 | M4 (10c) | 16 GB | llama-3.1-8b-instruct | 4bit | 1k | 236.2 | 20.5 | 26-03-10 |
| 42612 | M3 (10c) | 16 GB | Qwen3.5-4B | 4bit | 1k | 309.5 | 33.4 | 26-03-10 |
| 42613 | M4 Pro (16c) | 64 GB | Qwen3.5-9B | 4bit | 4k | 364.7 | 48.0 | 26-03-10 |
| 42614 | M4 Pro (16c) | 64 GB | Qwen3.5-9B | 4bit | 1k | 361.5 | 49.5 | 26-03-10 |
| 42615 | M3 Ultra (60c) | 256 GB | Qwen3.5-35B-A3B | 4bit | 32k | 1,675 | 68.2 | 26-03-10 |
| 42616 | M3 Ultra (60c) | 256 GB | Qwen3.5-35B-A3B | 4bit | 64k | 1,290 | 50.8 | 26-03-10 |
| 42617 | M3 Ultra (60c) | 256 GB | Qwen3.5-35B-A3B | 4bit | 8k | 2,069 | 85.2 | 26-03-10 |
| 42618 | M3 Ultra (60c) | 256 GB | Qwen3.5-35B-A3B | 4bit | 16k | 1,921 | 77.6 | 26-03-10 |
| 42619 | M3 Ultra (60c) | 256 GB | Qwen3.5-35B-A3B | 4bit | 1k | 2,045 | 87.8 | 26-03-10 |
| 42620 | M3 Ultra (60c) | 256 GB | Qwen3.5-35B-A3B | 4bit | 4k | 2,134 | 85.8 | 26-03-10 |