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 ↓ |
|---|---|---|---|---|---|---|---|---|
| 336091 | M3 Ultra (60c) | 256 GB | Qwen3.5-35B-A3B | 4bit | 1k | 2,045 | 87.8 | 26-03-10 |
| 336092 | M3 Ultra (60c) | 256 GB | Qwen3.5-35B-A3B | 4bit | 4k | 2,134 | 85.8 | 26-03-10 |
| 336093 | M4 (10c) | 16 GB | Qwen3.5-4B | 4bit | 8k | 371.4 | 26.7 | 26-03-10 |
| 336094 | M4 (10c) | 16 GB | Qwen3.5-4B | 4bit | 4k | 376.2 | 28.9 | 26-03-10 |
| 336095 | M4 (10c) | 16 GB | Qwen3.5-4B | 4bit | 1k | 367.9 | 30.4 | 26-03-10 |
| 336096 | M1 Max (32c) | 64 GB | Qwen3.5-27B | 4bit | 1k | 90.0 | 19.1 | 26-03-10 |
| 336097 | M1 Max (32c) | 64 GB | Qwen3.5-27B | 4bit | 4k | 90.4 | 18.0 | 26-03-10 |
| 336098 | M4 Max (40c) | 64 GB | NVIDIA-Nemotron-3-Nano-30... | 4bit | 64k | 1,062 | 40.7 | 26-03-10 |
| 336099 | M2 Max (38c) | 64 GB | Qwen3.5-4B | 8bit | 4k | 347.6 | 34.3 | 26-03-10 |
| 336100 | M2 Max (38c) | 64 GB | Qwen3.5-4B | 8bit | 1k | 413.9 | 26.9 | 26-03-10 |