Benchmark result
unsloth/Qwen3.8-27B-GGUF on NVIDIA GeForce RTX 4070 — 41.0 tok/s
Measured with Unsloth Studio on September 27, 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.
How this run compares
4th fastest of 8 runs of this model on NVIDIA GeForce RTX 4070 · median 40.7 tok/s.
sort
- 142.1tok/s76.0llama.cpp
- 241.7tok/s51.7Unsloth Studio · UD-Q4_K_XL · KV q8_0
- 341.6tok/s56.9Unsloth Studio · UD-Q4_K_XL · KV f16
- 441.0tok/s43.2Unsloth Studio · UD-Q4_K_XL · KV f16this run
- 540.3tok/s30.6llama.cpp
- 639.8tok/s43.6llama.cpp
- 739.3tok/s63.4llama.cpp
- 834.8tok/s66.4llama.cpp
unsloth/Qwen3.8-27B-GGUF on other hardware
Reproduce this run
toolUnsloth Studio
modelunsloth/Qwen3.8-27B-GGUF (UD-Q4_K_XL)
context16,384
kv cachef16
thinkingon
effortxhigh (model default)
think limitnone
samplingtemperature 0.7 · top_p 0.8 · top_k 20 · min_p 0 · presence_penalty 1.5
clientv0.4.80+98
Set those in Unsloth Studio, then:
llm-benchmark benchmark --model "unsloth/Qwen3.8-27B-GGUF" --tool "Unsloth Studio"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.