Benchmark result
qwen3.6:35b on Apple M5 Max — 19.6 tok/s
Measured with Ollama 0.23.2 on May 16, 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
1st fastest of 3 runs of this model on Apple M5 Max · median 15.0 tok/s.
sort
- 119.6tok/s77.0Ollama · Q4_K_Mthis run
- 215.0tok/s77.8Ollama · Q4_K_M
- 314.7tok/s77.8Ollama · Q4_K_M
Reproduce this run
toolOllama 0.23.2
modelqwen3.6:35b (Q4_K_M)
context65,536
clientv0.4.22+97
Set those in Ollama, then:
llm-benchmark benchmark --model "qwen3.6:35b" --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.