Benchmark result

Qwen3.8-9B-Q8_0 on 2× NVIDIA GeForce RTX 5060 — 73.5 tok/s

Measured with llama.cpp b10851-67672dc5b on September 8, 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 2× NVIDIA GeForce RTX 5060 · Multi-GPU · median 71.2 tok/s.

sort

173.5tok/s75.9llama.cpp · Q8_0this run
269.0tok/s77.6llama.cpp · Q8_0

Compare the nearest 5 →

Qwen3.8-9B-Q8_0 on other hardware

Reproduce this run

toolllama.cpp b10851-67672dc5b

modelQwen3.8-9B-Q8_0 (Q8_0)

context131,072

thinkingon

samplingtemperature 0.6000000238418579 · top_p 0.8999999761581421 · top_k 0 · min_p 0.05000000074505806 · presence_penalty 0 · repeat_penalty 1

clientv0.4.55+97

Set those in llama.cpp, then:

llm-benchmark benchmark --model "Qwen3.8-9B-Q8_0" --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.