Benchmark result
HermesFableAgentThink on Apple M5 Max — 15.9 tok/s
Measured with oMLX 0.32.0 on September 27, 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 4 runs of this model on Apple M5 Max · median 20.6 tok/s.
sort
- 121.7tok/s86.0oMLX
- 221.0tok/s78.2oMLX
- 320.2tok/s79.6oMLX
- 415.9tok/s82.3oMLXthis run
Reproduce this run
tooloMLX 0.32.0
modelHermesFableAgentThink
context262,144
thinkingon
samplingtemperature 1 · top_p 0.95 · top_k 20 · repetition_penalty 1
clientv0.4.80+98
Set those in oMLX, then:
llm-benchmark benchmark --model "HermesFableAgentThink" --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.