Benchmark result
Llama-3.1-8B-Instruct-4bit on Apple M5 Max — 94.2 tok/s
Measured with oMLX 0.7.1.dev1 on October 13, 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.
Reproduce this run
tooloMLX 0.7.1.dev1
modelLlama-3.1-8B-Instruct-4bit (4bit)
context8,192
kv cachenone
think limitnone
samplingtemperature 1 · top_p 0.95 · top_k 0 · repetition_penalty 1
clientv0.4.92+102
Set those in oMLX, then:
llm-benchmark benchmark --model "Llama-3.1-8B-Instruct-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.