Benchmark result
qwen/qwen3-coder-30b on Apple M5 Pro — 84.5 tok/s
Measured with LM Studio on August 28, 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
1st fastest of 2 runs of this model on Apple M5 Pro · median 83.7 tok/s.
sort
- 184.5tok/s54.4LM Studiothis run
- 282.9tok/s65.4Ollama · Q4_K_M
qwen/qwen3-coder-30b on other hardware
Reproduce this run
toolLM Studio
modelqwen/qwen3-coder-30b
context32,768
clientv0.4.44+97
Set those in LM Studio, then:
llm-benchmark benchmark --model "qwen/qwen3-coder-30b" --tool "LM Studio"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.