Benchmark result
SiliconSpecies/Swift-1.5-Qwen3.8-27B-Splash on Apple M5 Max — 50.8 tok/s
Measured with vLLM on September 27, 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.
Reproduce this run
toolvLLM
modelSiliconSpecies/Swift-1.5-Qwen3.8-27B-Splash
context262,144
thinkingon
clientv0.4.80+98
Set those in vLLM, then:
llm-benchmark benchmark --model "SiliconSpecies/Swift-1.5-Qwen3.8-27B-Splash" --tool "vLLM"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.