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 ↓
17861 M4 (10c) 24 GB Llama-3.2-3B-Instruct 4bit 4k 478.6 37.0 26-03-09
17862 M4 (10c) 24 GB Qwen3.5-4B 4bit 4k 380.6 34.1 26-03-09
17863 M4 (10c) 24 GB Qwen3.5-4B 4bit 1k 378.3 35.1 26-03-09
17864 M2 Ultra (76c) 128 GB Qwen3.5-0.8B bf16 64k 4,660 83.2 26-03-09
17865 M4 (10c) 16 GB Qwen3.5-4B 4bit 1k 400.3 38.8 26-03-09
17866 M4 (10c) 16 GB Qwen3.5-0.8B 8bit 1k 2,051 99.8 26-03-09
17867 M4 Max (40c) 128 GB Qwen3.5-9B 8bit 4k 890.4 50.9 26-03-09
17868 M4 Max (40c) 128 GB Qwen3.5-9B 8bit 1k 878.9 52.1 26-03-09
17869 M4 (10c) 16 GB Qwen3.5-0.8B 8bit 1k 1,656 90.4 26-03-09
17870 M4 (10c) 16 GB Qwen3.5-4B 4bit 4k 327.4 29.8 26-03-09
18,069 results · Page 1787 of 1807