Benchmark result
Tiel-Coder-35B-A3B-MLX-oQ4e-MTP on Apple M4 Max — 149.9 tok/s
Measured with oMLX manual on October 20, 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
2nd fastest of 3 runs of this model on Apple M4 Max · median 149.9 tok/s.
sort
- 1165.2tok/s61.4oMLX · oQ4e · KV none
- 2149.9tok/s82.3oMLX · oQ4e · KV nonethis run
- 3144.2tok/s71.6oMLX · oQ4e · KV none
Reproduce this run
tooloMLX manual
modelTiel-Coder-35B-A3B-MLX-oQ4e-MTP (oQ4e)
context262,144
kv cachenone
thinkingon
think limitnone
samplingtemperature 1 · top_p 0.95 · top_k 20 · repetition_penalty 1
clientv0.4.96+102
Set those in oMLX, then:
llm-benchmark benchmark --model "Tiel-Coder-35B-A3B-MLX-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.