# potamides/DeTikZify

Synthesizing Graphics Programs for Scientific Figures and Sketches with TikZ.

Repository: https://github.com/potamides/DeTikZify
Canonical: https://ross.abutalabs.com/products/detikzify
Homepage: https://nllg-detikzify.hf.space
Language: Python
License: Apache-2.0
License Family: permissive
Topics: huggingface, inverse-graphics, latex, llama, llm, multimodal, tikz, transformers, draw, graph, sketch, vectorization, visualization
Last push: 2026-02-04T15:01:54+00:00

## Health v2 (maintenance only)
Score: 49/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 65, release rhythm 16, longevity 72
- inputs: {"age_days": 1010, "days_push": 210, "days_rel": 422, "gap_med": 104, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1817, forks 92 (observed 2026-08-28T04:05:39.971357+00:00)

## What it is
DeTikZify is a Python library and research tool that uses multimodal large language models to synthesize TikZ/LaTeX graphics programs from scientific figures and sketches. It converts raster images of figures into editable vector TikZ code.

## Use cases
- convert a scientific figure image into TikZ code
- recreate a sketch as editable LaTeX graphics
- vectorize a plot into TikZ programmatically
- generate LaTeX figures from images with an LLM
- reverse-engineer graphics into TikZ programs
- edit a figure by regenerating its TikZ source

## When to choose
- you need editable TikZ/LaTeX source for an existing figure
- you work with scientific figures and want vector reproduction
- you want to experiment with multimodal LLMs for inverse graphics

## When to avoid
- you need general-purpose image vectorization without TikZ output
- you have no GPU or LLM inference resources
- you need pixel-perfect raster-to-vector tracing

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, llm-inference, image-processing, data-visualization, graphics
- domain: machine-learning, large-language-models, computer-vision, graphics, artificial-intelligence
- platform: python, cross-platform
- tags: tikz, latex, inverse-graphics, multimodal, llama, huggingface, sketches, scientific-figures, vectorization, gpu

## Member repositories
- potamides/DeTikZify (main) score 49

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:39.971357+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:20:14.007977+00:00, confidence not recorded.
  - readme: https://github.com/potamides/DeTikZify (fetched 2026-08-28T04:05:39.971357+00:00, sha 5535a72aa6f1)
- Data as of 2026-08-30T08:39:29.467469+00:00.