Benchmark result
ulkaa/Qwen3.8-27B-AWQ-INT4 — 185.0 tok/s
Measured with vLLM on September 8, 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
ulkaa/Qwen3.8-27B-AWQ-INT4 on other hardware
- 185.0tok/s86.5Unknown hardware · INT4this run
- 37.2tok/s89.6NVIDIA GeForce RTX 3080 +1 more · INT4
Reproduce this run
toolvLLM
modelulkaa/Qwen3.8-27B-AWQ-INT4 (INT4)
context65,536
clientv0.4.55+97
Set those in vLLM, then:
llm-benchmark benchmark --model "ulkaa/Qwen3.8-27B-AWQ-INT4" --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.