Benchmark result
unsloth/Qwen3.8-27B-GGUF:UD-Q4_K_M on Apple M4 Pro — 37.5 tok/s
Measured with Splash manual on October 15, 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
4th fastest of 4 runs of this model on Apple M4 Pro · median 38.0 tok/s.
sort
- 138.6tok/s62.8Splash · Q4_K_M
- 238.3tok/s82.1Splash · Q4_K_M
- 337.6tok/s79.9Splash · Q4_K_M
- 437.5tok/s85.5Splash · Q4_K_Mthis run
unsloth/Qwen3.8-27B-GGUF:UD-Q4_K_M on other hardware
Reproduce this run
toolSplash manual
modelunsloth/Qwen3.8-27B-GGUF:UD-Q4_K_M (Q4_K_M)
context262,144
thinkingon
clientv0.4.93+102
Set those in Splash, then:
llm-benchmark benchmark --model "unsloth/Qwen3.8-27B-GGUF:UD-Q4_K_M" --tool "Splash"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.