Benchmark result
Qwen3.6-35B-A3B-MLX-8bit on Apple M5 Max — 68.4 tok/s
Measured with oMLX on June 2, 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
4th fastest of 6 runs of this model on Apple M5 Max · median 68.8 tok/s.
sort
- 199.4tok/s86.0oMLX
- 290.1tok/s83.1oMLX
- 369.2tok/s77.7oMLX
- 468.4tok/s78.2oMLXthis run
- 562.6tok/s74.6oMLX
- 624.7tok/s80.2oMLX
Reproduce this run
tooloMLX
modelQwen3.6-35B-A3B-MLX-8bit
context65,536
clientv0.4.22+97
Set those in oMLX, then:
llm-benchmark benchmark --model "Qwen3.6-35B-A3B-MLX-8bit" --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.