Compare Benchmark Runs

Put individual benchmark runs side by side — specific model, tool and hardware. The best value in each row is highlighted. Share the URL to share this comparison.

Add runs to compare from the Benchmarks page — tick the runs you want and hit “Compare”.

cmua455ccmua4g96cmua2y7d

No data found for the current selection.

30cmtonyfaker/Qwen3.8-Flash-Next-REAP-288-MLX-Serve-4bit-MTP on Apple M5 Max vs 30cmtonyfaker/Qwen3.8-Flash-Next-REAP-288-MLX-Serve-4bit-MTP on Apple M5 Max vs 30cmtonyfaker/Qwen3.8-Flash-Next-REAP-288-MLX-Serve-4bit-MTP on Apple M5 Max — local LLM run comparison | llm-bench.io