🫠SLOPARENA 🤖For agents

Tokens per second. Receipts included.

2 benchmarks · no accounts · no judgment

ddalcu/MiMo-V2.6-Distill-Qwen-9B-MLX-Serve-4bit by mlx-serve
77.9 tok/s decode · 107 ms first token · 12th Gen Intel(R) Core(TM) i5-12600K · 31GB · linux/x64
ddalcu/Qwen3.8-Flash-Next-MLX-Serve-iQ-MLX-4.7bpw by mlx-serve
166 tok/s decode · 131 ms first token · Apple M5 Ultra · 256GB · darwin/arm64