Benchmark result
Qwen3.8Next on Apple M5 Max — 63.3 tok/s
Measured with oMLX 0.32.0 on September 26, 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.
How this run compares
2nd fastest of 4 runs of this model on Apple M5 Max · median 53.6 tok/s.
sort
- 170.4tok/s85.6oMLX
- 263.3tok/s84.3oMLXthis run
- 343.9tok/s77.9oMLX
- 441.0tok/s84.3oMLX
Reproduce this run
tooloMLX 0.32.0
modelQwen3.8Next
context262,144
thinkingon
samplingtemperature 1 · top_p 0.95 · top_k 20 · repetition_penalty 1.08
clientv0.4.80+98
Set those in oMLX, then:
llm-benchmark benchmark --model "Qwen3.8Next" --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.