Benchmark result
deepseek/deepseek-r1-0528-qwen3-8b — 31.9 tok/s
Measured with LM Studio on September 5, 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
deepseek/deepseek-r1-0528-qwen3-8b on other hardware
- 42.8tok/s32.4Apple M4 Max
- 31.9tok/s28.8Unknown hardwarethis run
Reproduce this run
toolLM Studio
modeldeepseek/deepseek-r1-0528-qwen3-8b
context8,192
clientv0.4.55+97
Set those in LM Studio, then:
llm-benchmark benchmark --model "deepseek/deepseek-r1-0528-qwen3-8b" --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.