Benchmark result
Qwen3.8Next2 on Apple M5 Max — 48.8 tok/s
Measured with oMLX 0.32.0 on September 20, 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 0.32.0
modelQwen3.8Next2
context262,144
samplingtemperature 1 · top_p 0.95 · top_k 20 · repetition_penalty 1.08
clientv0.4.71+98
Set those in oMLX, then:
llm-benchmark benchmark --model "Qwen3.8Next2" --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.