Benchmark result
Qwopus3.8-27B-Flash-V2-oQ4e-mtp on Apple M5 Max — 65.1 tok/s
Measured with oMLX 0.31.4.dev132+g94cdcae13 on October 11, 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.
How this run compares
3rd fastest of 3 runs of this model on Apple M5 Max · median 67.4 tok/s.
sort
- 168.1tok/s76.4oMLX · oQ4e · KV none
- 267.4tok/s69.0oMLX · oQ4e · KV none
- 365.1tok/s77.5oMLX · oQ4e · KV nonethis run
Reproduce this run
tooloMLX 0.31.4.dev132+g94cdcae13
modelQwopus3.8-27B-Flash-V2-oQ4e-mtp (oQ4e)
context65,536
kv cachenone
effortxhigh (model default)
think limitnone
samplingtemperature 1 · top_p 0.95 · top_k 0 · repetition_penalty 1
clientv0.4.89+102
Set those in oMLX, then:
llm-benchmark benchmark --model "Qwopus3.8-27B-Flash-V2-oQ4e-mtp" --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.