v0.3.0

What's New

Introducing DeTikZify_v2.5

We release DeTikZify_v2.5 (8b), our latest model, which builds on DeTikZify_v2 using reinforment learning from self-feedback (RLSF). We make it the new default model in our Hugging Face Space. Check out the model card for more information.

Support for Windows

Inference now works on Linux, macOS, and Windows operating systems.

Document Docker Use

The DeTikZify Space can be run locally with Docker.

v0.2.1

Release accompanying TikZero adapters which plug directly into DeTikZify_v2 (8b) and enable zero-shot text-conditioning, and TikZero+ with additional end-to-end fine-tuning. Check out our paper for more information.

v0.2.0

Release accompanying DeTikZify_v2 (8b), our latest model which surpasses all previous DeTikZify versions in our evaluation. Check out the model card for more information.

v0.1.0

Initial release of DeTikZify, together with the paper DeTikZify: Synthesizing Graphics Programs for Scientific Figures and Sketches with TikZ.