Benchmark result
Qwen3.8-27b-q4s on NVIDIA GeForce RTX 4090 D — 67.3 tok/s
Measured with llama.cpp b2-991c2b8 on October 30, 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 2 runs of this model on NVIDIA GeForce RTX 4090 D · median 65.7 tok/s.
sort
- 167.3tok/s89.6llama.cppthis run
- 264.2tok/s76.6llama.cpp
Reproduce this run
toolllama.cpp b2-991c2b8
modelQwen3.8-27b-q4s
context200,000
thinkingon
effortmedium (model default)
think limitnone
samplingtemperature 1 · top_p 0.949999988079071 · top_k 20 · min_p 0.05000000074505806 · presence_penalty 0 · repeat_penalty 1
clientv0.4.98+102
Set those in llama.cpp, then:
llm-benchmark benchmark --model "Qwen3.8-27b-q4s" --tool "llama.cpp"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.