Benchmark result
Ornith-1.5-35B-A3B-MLX-4bit on Apple M4 Pro — 82.6 tok/s
Measured with oMLX manual on October 15, 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 4 runs of this model on Apple M4 Pro · median 83.3 tok/s.
sort
- 184.6tok/s75.0oMLX · KV none
- 284.1tok/s69.5oMLX · 4bit · KV none
- 382.6tok/s69.9oMLX · 4bit · KV nonethis run
- 481.6tok/s83.6oMLX · 4bit · KV none
Reproduce this run
tooloMLX manual
modelOrnith-1.5-35B-A3B-MLX-4bit (4bit)
context16,384
kv cachenone
thinkingon
think limit4,096 tokens
samplingtemperature 1 · top_p 0.95 · top_k 20 · repetition_penalty 1
clientv0.4.93+102
Set those in oMLX, then:
llm-benchmark benchmark --model "Ornith-1.5-35B-A3B-MLX-4bit" --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.