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 ↓
154701 M5 Pro (20c) 64 GB NVIDIA-Nemotron-3-Nano-30... 8bit 32k 1,801 46.3 26-04-26
154702 M5 Pro (20c) 64 GB NVIDIA-Nemotron-3-Nano-30... 8bit 16k 1,893 58.3 26-04-26
154703 M5 Pro (20c) 64 GB NVIDIA-Nemotron-3-Nano-30... 8bit 8k 1,900 64.6 26-04-26
154704 M5 Pro (20c) 64 GB NVIDIA-Nemotron-3-Nano-30... 8bit 4k 1,852 66.2 26-04-26
154705 M5 Pro (20c) 64 GB NVIDIA-Nemotron-3-Nano-30... 8bit 1k 1,362 66.9 26-04-26
154706 M5 Max (40c) 128 GB Qwen3.6-35B-A3B 8bit 8k 3,910 93.9 26-04-26
154707 M5 Max (40c) 128 GB Qwen3.6-35B-A3B 8bit 64k 4,797 45.0 26-04-26
154708 M5 Max (40c) 128 GB Qwen3.6-35B-A3B 8bit 128k 3,365 39.4 26-04-26
154709 M5 Max (40c) 128 GB Qwen3.6-35B-A3B 8bit 1k 2,953 264.7 26-04-26
154710 M5 Max (40c) 128 GB Qwen3.6-35B-A3B 8bit 4k 1,006 96.5 26-04-26
295,181 results · Page 15471 of 29519