Benchmark result
Ornith-1.5-35B-A3B-MLX-8bit on Apple M5 Max — 100.7 tok/s
Measured with oMLX 0.6.2 on August 22, 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
1st fastest of 2 runs of this model on Apple M5 Max · median 96.7 tok/s.
sort
- 1100.7tok/s81.4oMLXthis run
- 292.6tok/s82.5oMLX
Reproduce this run
tooloMLX 0.6.2
modelOrnith-1.5-35B-A3B-MLX-8bit
context262,144
thinkingon
clientv0.4.40+97
Set those in oMLX, then:
llm-benchmark benchmark --model "Ornith-1.5-35B-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.