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 ↓ |
|---|---|---|---|---|---|---|---|---|
| 206451 | M5 (10c) | 32 GB | gemma-4-31b-it | 4bit | 4k | 152.0 | 6.4 | 26-04-16 |
| 206452 | M5 (10c) | 32 GB | gemma-4-31b-it | 4bit | 1k | 159.1 | 6.9 | 26-04-16 |
| 206453 | M2 Max (38c) | 32 GB | Qwen3.5-27B-Claude-4.6-Op... | 4bit | 4k | 114.6 | 19.9 | 26-04-16 |
| 206454 | M2 Max (38c) | 32 GB | Qwen3.5-27B-Claude-4.6-Op... | 4bit | 1k | 112.5 | 20.6 | 26-04-16 |
| 206455 | M3 Max (30c) | 36 GB | Qwen3-Coder-30B-A3B-Instr... | 4bit | 8k | 878.6 | 52.0 | 26-04-16 |
| 206456 | M4 (10c) | 32 GB | Gemma-4-E2B | 4bit | 4k | 2,079 | 63.3 | 26-04-16 |
| 206457 | M4 (10c) | 32 GB | Gemma-4-E2B | 4bit | 1k | 1,552 | 66.6 | 26-04-16 |
| 206458 | M3 Max (30c) | 36 GB | Qwen3-Coder-30B-A3B-Instr... | 4bit | 4k | 1,021 | 64.7 | 26-04-16 |
| 206459 | M3 Max (30c) | 36 GB | Qwen3-Coder-30B-A3B-Instr... | 4bit | 1k | 966.3 | 80.4 | 26-04-16 |
| 206460 | M1 Ultra (64c) | 128 GB | GLM-4.7-Flash | 8bit | 4k | 749.3 | 38.6 | 26-04-16 |