Benchmark result
qwen2.5-coder-32b-instruct-q3_k_m on 3× NVIDIA GeForce RTX 5070 Ti — 6.7 tok/s
Measured with llama.cpp b11029-5c53396b8 on September 18, 2026.
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.
Reproduce this run
toolllama.cpp b11029-5c53396b8
modelqwen2.5-coder-32b-instruct-q3_k_m (Q3_K_M)
context8,192
samplingtemperature 0.800000011920929 · top_p 0.949999988079071 · top_k 40 · min_p 0.05000000074505806 · presence_penalty 0 · repeat_penalty 1
clientv0.4.59+97
Set those in llama.cpp, then:
llm-benchmark benchmark --model "qwen2.5-coder-32b-instruct-q3_k_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.