Benchmark result
hf.co/unsloth/Qwen3.8-27B-GGUF:UD-IQ3_S on NVIDIA GeForce RTX 4080 SUPER — 46.7 tok/s
Measured with Ollama 0.34.1 on September 19, 2026.
How this run compares
2nd fastest of 2 runs of this model on NVIDIA GeForce RTX 4080 SUPER · median 46.7 tok/s.
sort
- 146.7tok/s41.2llama.cpp
- 246.7tok/s67.3Ollama · IQ3_Sthis run
Reproduce this run
toolOllama 0.34.1
modelhf.co/unsloth/Qwen3.8-27B-GGUF:UD-IQ3_S (IQ3_S)
context32,768
effortxhigh (model default)
clientv0.4.59+97
Set those in Ollama, then:
llm-benchmark benchmark --model "hf.co/unsloth/Qwen3.8-27B-GGUF:UD-IQ3_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.