Benchmark result
qwen3:8b on NVIDIA GeForce RTX 3060 Ti — 59.5 tok/s
Measured with Ollama 0.34.1 on September 23, 2026.
How this run compares
qwen3:8b on other hardware
Reproduce this run
toolOllama 0.34.1
modelqwen3:8b (Q4_K_M)
context8,192
thinkingon
samplingtemperature 0.6 · top_p 0.95 · top_k 20 · repeat_penalty 1
clientv0.4.80+98
Set those in Ollama, then:
llm-benchmark benchmark --model "qwen3:8b" --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.