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 ↓
20051 M4 Max (40c) 64 GB Qwen3.5-9B 4bit 4k 883.5 84.5 26-03-09
20052 M4 Max (40c) 64 GB Qwen3.5-9B 4bit 1k 877.9 87.1 26-03-09
20053 M1 Pro (16c) 32 GB Qwen3.5-9B 6bit 4k 155.9 23.2 26-03-09
20054 M1 Pro (16c) 32 GB Qwen3.5-9B 6bit 1k 155.3 24.2 26-03-09
20055 M4 Max (40c) 64 GB Qwen3.5-35B-A3B 4bit 4k 1,700 107.6 26-03-09
20056 M4 Max (40c) 64 GB Qwen3.5-35B-A3B 4bit 1k 1,487 113.8 26-03-09
20057 M4 (8c) 16 GB Qwen3.5-4B 4bit 4k 299.5 34.6 26-03-09
20058 M4 (8c) 16 GB Qwen3.5-4B 4bit 1k 297.1 35.6 26-03-09
20059 M4 (10c) 16 GB Qwen3.5-9B 4bit 1k 197.8 21.3 26-03-09
20060 M3 Ultra (60c) 96 GB Qwen3.5-35B-A3B9bit 8bit 32k 1,620 51.3 26-03-09
20,820 results · Page 2006 of 2082