# BIT-DataLab/Edit-Banana

Edit Banana: A framework for converting statistical formats into editable.

Repository: https://github.com/BIT-DataLab/Edit-Banana
Canonical: https://ross.abutalabs.com/products/edit-banana
Homepage: https://www.editbanana.net
Language: Python
License: AGPL-3.0
License Family: copyleft
Topics: ai, data, figure, llm, nanobanana, open-source, python, pythonprogramming
Last push: 2026-08-25T04:53:14+00:00

## Health v2 (maintenance only)
Score: 60/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 35, longevity 16
- inputs: {"age_days": 229, "days_push": 8, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 5469, forks 359 (observed 2026-08-28T04:09:19.608432+00:00)

## What it is
Edit Banana is an open-source Python framework that converts static images and PDFs of diagrams, flowcharts, and charts into fully editable DrawIO (XML) and PPTX files. It uses a fine-tuned SAM 3 segmentation model combined with multimodal LLMs (Qwen/GPT-4V) and OCR to reconstruct layout, shapes, arrows, and text with high fidelity.

## Use cases
- convert a flowchart image into an editable drawio file
- turn a screenshot of an architecture diagram into editable shapes
- extract diagrams from pdfs into pptx
- reconstruct chart elements so I can edit text and colors
- make an uneditable figure editable
- ocr text inside diagram images and export as xml

## When to choose
- you need to edit diagrams that only exist as images or PDFs
- you want 1:1 preservation of layout, colors, strokes, and arrow styles
- you prefer an open-source, self-hostable pipeline over manual redrawing

## When to avoid
- you need the very latest features - the GitHub repo trails the hosted web service
- you lack a CUDA-capable GPU for local inference
- you only need simple image editing rather than structured diagram reconstruction

## Facets
- artifact type: framework
- maturity: active
- function: ocr, image-processing, machine-learning, llm-inference, parser, data-visualization
- domain: computer-vision, artificial-intelligence, large-language-models, data-visualization
- platform: python, cross-platform
- tags: image-to-diagram, drawio, sam3, vlm, diagram-reconstruction, pptx-export, editable-assets, diagrams, gpu, web-server

## Member repositories
- BIT-DataLab/Edit-Banana (main) score 60

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:19.608432+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 2026-08-29T17:56:44.483222+00:00, confidence not recorded.
  - readme: https://github.com/BIT-DataLab/Edit-Banana (fetched 2026-08-28T04:09:19.608432+00:00, sha a977e80b576d)
  - homepage: https://www.editbanana.net (fetched 2026-08-29T08:51:47.960545+00:00, sha a89fdf6136d3)
- Data as of 2026-08-30T08:39:29.467469+00:00.