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 ↓ |
|---|---|---|---|---|---|---|---|---|
| 2031 | M2 Ultra (76c) | 192 GB | qwen3.5-9bmxfp8 | 8bit | 1k | 1,080 | 63.5 | 26-03-09 |
| 2032 | M1 (7c) | 16 GB | Qwen3.5-4B | 4bit | 4k | 124.2 | 19.9 | 26-03-09 |
| 2033 | M1 (7c) | 16 GB | Qwen3.5-4B | 4bit | 1k | 124.7 | 20.5 | 26-03-09 |
| 2034 | M4 (10c) | 16 GB | Qwen3.5-9B | 4bit | 4k | 222.5 | 19.7 | 26-03-09 |
| 2035 | M4 (10c) | 16 GB | Qwen3.5-9B | 4bit | 1k | 221.8 | 19.9 | 26-03-09 |
| 2036 | M4 Max (40c) | 128 GB | Qwen3.5-122B-A10B | 4bit | 4k | 583.9 | 54.0 | 26-03-09 |
| 2037 | M4 Max (40c) | 128 GB | Qwen3.5-122B-A10B | 4bit | 1k | 567.4 | 56.3 | 26-03-09 |
| 2038 | M4 (10c) | 16 GB | Qwen3.5-9B | 4bit | 1k | 223.6 | 20.8 | 26-03-09 |
| 2039 | M4 (10c) | 16 GB | Qwen3.5-9B | 4bit | 4k | 224.7 | 20.3 | 26-03-09 |
| 2040 | M1 Max (32c) | 32 GB | Qwen3.5-35B-A3B5.5bit | 5bit | 4k | 540.6 | 53.5 | 26-03-09 |