Benchmark result
gemma-4-31B-it-MLX-8bit on Apple M5 Max — 14.9 tok/s
Measured with oMLX 0.6.0.dev1 on August 14, 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
1st fastest of 2 runs of this model on Apple M5 Max · median 14.9 tok/s.
sort
- 114.9tok/s70.9oMLXthis run
- 214.9tok/s75.0oMLX
gemma-4-31B-it-MLX-8bit on other hardware
Reproduce this run
tooloMLX 0.6.0.dev1
modelgemma-4-31B-it-MLX-8bit
context65,536
clientv0.4.36+97
Set those in oMLX, then:
llm-benchmark benchmark --model "gemma-4-31B-it-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.