Benchmark result

Qwen3.8-27B-UD-IQ4_XS on AMD Radeon RX 9070/9070 XT/9070 GRE — 30.8 tok/s

Measured with vLLM on August 28, 2026.

What the model built

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

2nd fastest of 2 runs of this model on AMD Radeon RX 9070/9070 XT/9070 GRE · median 30.8 tok/s.

sort

130.9tok/s83.9vLLM
230.8tok/s84.3vLLMthis run

Compare the nearest 5 →

Qwen3.8-27B-UD-IQ4_XS on other hardware

Reproduce this run

toolvLLM

modelQwen3.8-27B-UD-IQ4_XS

context49,152

thinkingon

clientv0.4.44+97

Set those in vLLM, then:

llm-benchmark benchmark --model "Qwen3.8-27B-UD-IQ4_XS" --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.