Benchmark result
Qwen3.8-27B-GSQ-RCO-IQ3_S-mtp on AMD Radeon RX 7800 XT — 48.6 tok/s
Measured with qwen-mtp-think manual on November 16, 2026.
Support llm-bench.io
llm-bench.io is free. If it helped you, I'd appreciate your support to cover the infrastructure costs.
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
2nd fastest of 3 runs of this model on AMD Radeon RX 7800 XT · median 48.6 tok/s.
sort
- 153.1tok/s48.7rdna-fast
- 248.6tok/s80.1qwen-mtp-thinkthis run
- 326.5tok/s70.0llama.cpp
Qwen3.8-27B-GSQ-RCO-IQ3_S-mtp on other hardware
Reproduce this run
toolqwen-mtp-think manual
modelQwen3.8-27B-GSQ-RCO-IQ3_S-mtp
context8,192
thinkingon
clientv0.4.99+102
Set those in qwen-mtp-think, then:
llm-benchmark benchmark --model "Qwen3.8-27B-GSQ-RCO-IQ3_S-mtp" --tool "qwen-mtp-think"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.