Benchmark result
unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF:Q3_K_S on AMD Radeon RX 9070/9070 XT/9070 GRE — 107.7 tok/s
Measured with Ollama 0.35.1 on October 4, 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
toolOllama 0.35.1
modelunsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF:Q3_K_S (Q3_K_S)
context8,192
clientv0.4.87+102
Set those in Ollama, then:
llm-benchmark benchmark --model "unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF:Q3_K_S" --tool "Ollama"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.