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 ↓
15551 M3 (10c) 24 GB Llama-3.2-1B-Instruct 4bit 16k 898.6 57.1 26-03-09
15552 M3 (10c) 24 GB Llama-3.2-1B-Instruct 4bit 8k 1,130 59.3 26-03-09
15553 M3 (10c) 24 GB Llama-3.2-1B-Instruct 4bit 4k 1,216 76.7 26-03-09
15554 M3 (10c) 24 GB Llama-3.2-1B-Instruct 4bit 1k 1,249 102.4 26-03-09
15555 M4 (8c) 16 GB Llama-3.2-3B-Instruct 4bit 1k 419.8 47.9 26-03-09
15556 M4 (8c) 16 GB Llama-3.2-3B-Instruct 4bit 4k 382.4 37.8 26-03-09
15557 M2 Pro (16c) 16 GB Qwen3.5-9B 8bit 1k 171.5 20.0 26-03-09
15558 M2 Pro (16c) 16 GB Qwen3.5-9B 8bit 4k 172.3 18.1 26-03-09
15559 M3 Ultra (80c) 512 GB Qwen3.5-35B-A3B 4bit 1k 1,681 96.0 26-03-09
15560 M3 Ultra (80c) 512 GB Qwen3.5-35B-A3B 4bit 4k 2,425 92.2 26-03-09
16,634 results · Page 1556 of 1664