Benchmark result

qwen3.8-flash-next-coder-iq1_m on AMD Radeon RX 7900 XTX — 65.6 tok/s

Measured with llama.cpp Strata 0.1.38 on October 4, 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

qwen3.8-flash-next-coder-iq1_m on other hardware

43.3tok/s84.0AMD Radeon RX 9070/9070 XT/9070 GRE

Reproduce this run

toolllama.cpp Strata 0.1.38

modelqwen3.8-flash-next-coder-iq1_m

context65,536

thinkingon

clientv0.4.46+97

Set those in llama.cpp, then:

llm-benchmark benchmark --model "qwen3.8-flash-next-coder-iq1_m" --tool "llama.cpp"

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.