Benchmark result

qwen3.8-flash-next-iq3_s on AMD Radeon RX 9070/9070 XT/9070 GRE — 67.9 tok/s

Measured with Strata Strata 0.1.45 on October 18, 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

1st fastest of 2 runs of this model on AMD Radeon RX 9070/9070 XT/9070 GRE · median 66.4 tok/s.

sort

167.9tok/s83.6Stratathis run
264.9tok/s90.5Strata

Compare the nearest 5 →

qwen3.8-flash-next-iq3_s on other hardware

Reproduce this run

toolStrata Strata 0.1.45

modelqwen3.8-flash-next-iq3_s

context65,536

thinkingon

effortmedium

samplingtemperature 1 · top_p 0.95 · top_k 20

clientv0.4.96+102

Set those in Strata, then:

llm-benchmark benchmark --model "qwen3.8-flash-next-iq3_s" --tool "Strata"

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.