Hugging Face presents Palette, a 20B model for decks in a design language
🎧 Listen to this story
Hugging Face presents Palette as a bet on a small open-weight model that can run on a user’s own infrastructure and be trained to follow its own design language. The 20B model was fine-tuned with LoRA and synthetic data based on the IBM Design Language; the company says it ranked first in two of the three domains measured on PresentBench.
The system splits the work between a “designer,” which defines the structure, content and visual decisions in JSON, and a “coder,” which generates JavaScript for each slide. Hugging Face says the split reduces geometry errors such as clipped text and overlapping elements, allows individual slides to be edited and makes the coding stage about 6–7 times faster by enabling parallel calls. The company also says it used synthetic data generated by a larger open model because no public corpus combines presentation plans, design decisions and slide code.
Why it matters · editorial interpretation
Separating design decisions from code generation could make model-generated presentations more editable and reduce layout errors, according to Hugging Face. Using an open-weight model that can be trained on a user’s own infrastructure also enables adaptation to a specific visual language, without relying on a public corpus of this kind.