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 ↓ |
|---|---|---|---|---|---|---|---|---|
| 39051 | M3 Ultra (60c) | 256 GB | GLM-4.7-Flash | 6bit | 64k | 285.6 | 18.3 | 26-03-10 |
| 39052 | M3 Ultra (60c) | 256 GB | GLM-4.7-Flash | 6bit | 16k | 806.0 | 42.3 | 26-03-10 |
| 39053 | M3 Ultra (60c) | 256 GB | GLM-4.7-Flash | 6bit | 32k | 512.2 | 31.4 | 26-03-10 |
| 39054 | M3 Ultra (60c) | 256 GB | GLM-4.7-Flash | 6bit | 4k | 1,310 | 59.2 | 26-03-10 |
| 39055 | M3 Ultra (60c) | 256 GB | GLM-4.7-Flash | 6bit | 8k | 1,114 | 52.2 | 26-03-10 |
| 39056 | M3 Ultra (60c) | 256 GB | GLM-4.7-Flash | 6bit | 1k | 1,061 | 64.5 | 26-03-10 |
| 39057 | M2 (8c) | 16 GB | Qwen3.5-9B | 4bit | 4k | 86.8 | 17.9 | 26-03-10 |
| 39058 | M2 (8c) | 16 GB | Qwen3.5-9B | 4bit | 1k | 86.3 | 18.5 | 26-03-10 |
| 39059 | M1 Max (32c) | 64 GB | Qwen3.5-35B-A3B | 6bit | 64k | 325.6 | 20.3 | 26-03-10 |
| 39060 | M1 Max (32c) | 64 GB | Qwen3.5-35B-A3B | 6bit | 32k | 379.9 | 30.2 | 26-03-10 |