Benchmark result
KAT-Coder-V2.5-Dev-OptiQ-4bit-REAP-18B on Apple M4 Max — 56.7 tok/s
Measured with oMLX manual on October 22, 2026.
What the model built
Open full screen →The coding scenario asks for a playable game in a single HTML file. This is exactly what the model returned, unedited; the frame only adds a line that tells this page how tall it is.
Reproduce this run
tooloMLX manual
modelKAT-Coder-V2.5-Dev-OptiQ-4bit-REAP-18B (4bit)
context65,536
kv cachenone
thinkingoff
samplingtemperature 0.7 · top_p 0.8 · top_k 20 · min_p 0 · repetition_penalty 1
clientv0.4.96+102
Set those in oMLX, then:
llm-benchmark benchmark --model "KAT-Coder-V2.5-Dev-OptiQ-4bit-REAP-18B" --tool "oMLX"The client prompts for context length and thinking mode, and for the KV cache dtype on Unsloth Studio. Sampling is left at the model default — the values above are what the tool reported using, not overrides the client sent.