Benchmark result
qwen3:30b-thinking on Apple M5 Pro — 73.7 tok/s
Measured with Ollama 0.34.0 on September 12, 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 2 runs of this model on Apple M5 Pro · median 75.3 tok/s.
sort
- 176.8tok/s70.9Ollama · Q4_K_M
- 273.7tok/s70.8Ollama · Q4_K_Mthis run
Reproduce this run
toolOllama 0.34.0
modelqwen3:30b-thinking (Q4_K_M)
context65,536
thinkingon
samplingtemperature 0.6 · top_p 0.95 · top_k 20 · repeat_penalty 1
clientv0.4.59+97
Set those in Ollama, then:
llm-benchmark benchmark --model "qwen3:30b-thinking" --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.