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 ↓ |
|---|---|---|---|---|---|---|---|---|
| 2871 | M4 Max (40c) | 64 GB | NVIDIA-Nemotron-3-Nano-30... | 4bit | 32k | 1,135 | 99.3 | 26-03-10 |
| 2872 | M4 Max (40c) | 64 GB | NVIDIA-Nemotron-3-Nano-30... | 4bit | 4k | 1,519 | 136.4 | 26-03-10 |
| 2873 | M4 Max (40c) | 64 GB | NVIDIA-Nemotron-3-Nano-30... | 4bit | 8k | 1,475 | 131.8 | 26-03-10 |
| 2874 | M4 Max (40c) | 64 GB | NVIDIA-Nemotron-3-Nano-30... | 4bit | 1k | 1,200 | 141.8 | 26-03-10 |
| 2875 | M4 Max (40c) | 128 GB | Qwen3.5-35B-A3B | 8bit | 4k | 1,682 | 80.9 | 26-03-10 |
| 2876 | M4 Max (40c) | 128 GB | Qwen3.5-35B-A3B | 8bit | 1k | 1,448 | 85.2 | 26-03-10 |
| 2877 | M4 (10c) | 16 GB | Qwen3.5-2B | unknown | 16k | 740.5 | 5.2 | 26-03-10 |
| 2878 | M1 Pro (16c) | 32 GB | Qwen3.5-9B | 4bit | 4k | 149.9 | 29.8 | 26-03-10 |
| 2879 | M1 Pro (16c) | 32 GB | Qwen3.5-9B | 4bit | 1k | 149.7 | 31.1 | 26-03-10 |
| 2880 | M1 Pro (16c) | 16 GB | Qwen3.5-2B | 6bit | 4k | 615.1 | 62.7 | 26-03-10 |