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 ↓ |
|---|---|---|---|---|---|---|---|---|
| 156451 | M2 Ultra (76c) | 192 GB | Qwen3.6-35B-A3B-oQ4 | 4bit | 1k | 951.1 | 189.1 | 26-05-03 |
| 156452 | M2 Ultra (76c) | 192 GB | Qwen3.6-35B-A3B-oQ4 | 4bit | 4k | 644.6 | 76.2 | 26-05-03 |
| 156453 | M3 Ultra (60c) | 96 GB | Qwen2.5-7B-Instruct | 4bit | 8k | 1,225 | 101.0 | 26-05-03 |
| 156454 | M3 Ultra (60c) | 96 GB | Qwen2.5-7B-Instruct | 4bit | 4k | 1,243 | 114.2 | 26-05-03 |
| 156455 | M3 Ultra (80c) | 512 GB | gemma-4-31b-it | 8bit | 64k | 244.7 | 12.2 | 26-05-03 |
| 156456 | M5 (10c) | 24 GB | Qwen3.5-2B | 8bit | 4k | 1,198 | 28.9 | 26-05-03 |
| 156457 | M5 (10c) | 24 GB | Qwen3.5-2B | 8bit | 1k | 950.2 | 32.2 | 26-05-03 |
| 156458 | M4 (10c) | 24 GB | gpt-oss-20b-MXFP4-Q8 | 4bit | 4k | 393.3 | 32.7 | 26-05-03 |
| 156459 | M4 (10c) | 24 GB | gpt-oss-20b-MXFP4-Q8 | 4bit | 32k | 312.2 | 14.4 | 26-05-03 |
| 156460 | M3 Ultra (60c) | 96 GB | gemma-4-E4B-it | 8bit | 8k | 2,845 | 72.7 | 26-05-03 |