Benchmark result
qwen3.8-27b-turbo on 2× AMD Radeon AI PRO R9700 — 132.4 tok/s
Measured with radiance manual on November 14, 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
7th fastest of 16 runs of this model on 2× AMD Radeon AI PRO R9700 · Multi-GPU · median 131.3 tok/s.
sort
- 1154.1tok/s85.0radiance
- 2141.7tok/s87.2radiance
- 3134.2tok/s87.9radiance
- 4133.6tok/s88.4radiance
- 5133.2tok/s84.8radiance
- 6132.8tok/s84.2radiance
- 7132.4tok/s90.0radiancethis run
- 8131.6tok/s87.7radiance
- 9131.1tok/s86.4radiance
- 10130.7tok/s89.1radiance
- 11130.3tok/s87.1radiance
- 12128.3tok/s86.6radiance
- 13127.5tok/s89.0radiance
- 14127.0tok/s91.7radiance
- 15126.5tok/s89.0radiance
- 16119.7tok/s87.4radiance
Reproduce this run
toolradiance manual
modelqwen3.8-27b-turbo
context65,536
clientv0.4.99+102
Set those in radiance, then:
llm-benchmark benchmark --model "qwen3.8-27b-turbo" --tool "radiance"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.