Simon Willison tests Qwen 3.8 27B — a new Apache 2 licensed 27B-parameter vision model from Alibaba’s Qwen lab that fits in a 17GB GGUF file and runs on consumer hardware. The model is genuinely impressive, but its default configuration is actively harmful.
The problem: Qwen ships with reasoning effort set to xhigh. This turns trivial requests into epics:
- A “pelican riding a bicycle” SVG took 21 minutes and 22K reasoning tokens (vs 137 seconds with reasoning off)
- “Draw an SVG of a circle” produced an elaborate animated geometric study with concentric rings, tick marks, and a Bauhaus palette — beautiful, but entirely not what was asked
- When building a bounding box visualization tool, the model spent reasoning tokens deciding to draw its own pelican demo scene because the prompt used “pelicans” as the label
Willison’s advice is blunt: turn reasoning down or off. Without it, the model is fast and capable. With 15-30 tok/s from LM Studio (vs 74-184 from hosted APIs), speed is the main barrier to daily-driving this thing, but Multi-Token Prediction via llama.cpp gives a ~72% boost.
What makes this worth reading is not the benchmarks — it’s the demonstration. The model drives coding agents, writes correct bounding box coordinates, generates working tools from a single prompt, and does it all from a file smaller than most video games.
“The most important thing about Qwen 3.8 27B is what it demonstrates. We can have an open weights general purpose model with a long context, effective tool calling, strong vision ability, and competent code generation, and we can fit the whole thing in just a 17GB file.”
The recurring lesson: local models are closing in on “good enough” for serious work. The catch is that default configurations designed for benchmark scores make them feel worse than they are. Turn off the overthinking, and a capable daily-driver LLM lives on your laptop.