Benchmark result
qwen3.8-flash-next-iq3_xxs on NVIDIA GeForce RTX 4070 — 52.3 tok/s
Measured with llama.cpp PolyStrata 0.1.45 on November 15, 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 NVIDIA GeForce RTX 4070 · median 52.3 tok/s.
sort
- 154.0tok/s85.6Strata
- 252.3tok/s81.5llama.cppthis run
- 348.2tok/s87.7llama.cpp
Reproduce this run
toolllama.cpp PolyStrata 0.1.45
modelqwen3.8-flash-next-iq3_xxs
context32,768
thinkingon
effortmedium
think limitnone
samplingtemperature 1 · top_p 0.95 · top_k 20
clientv0.4.99+102
Set those in llama.cpp, then:
llm-benchmark benchmark --model "qwen3.8-flash-next-iq3_xxs" --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.