Benchmark result
qwen3.8:27b-mlx on Apple M5 Max — 29.9 tok/s
Measured with Ollama 0.32.12 on August 17, 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.
How this run compares
2nd fastest of 3 runs of this model on Apple M5 Max · median 29.9 tok/s.
sort
- 140.3tok/s83.8Ollama · nvfp4
- 229.9tok/s82.6Ollama · nvfp4this run
- 322.2tok/s67.5LM Studio
Reproduce this run
toolOllama 0.32.12
modelqwen3.8:27b-mlx (nvfp4)
context8,192
clientv0.4.36+97
Set those in Ollama, then:
llm-benchmark benchmark --model "qwen3.8:27b-mlx" --tool "Ollama"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.