llmsresearch

llmsresearch / paperbanana

Public

Open source implementation and extension of Google Research’s PaperBanana for automated academic figures, diagrams, and research visuals, expanded to new domains like slide generation.

academic-diagrams academic-research agentic-ai arxiv diagram-generation
1,083
154
100% credibility
Found Feb 05, 2026 at 36 stars 30x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

PaperBanana is an open-source framework that uses AI agents to automatically generate high-quality academic diagrams and statistical plots from natural language descriptions of research methods or data.

How It Works

1
📚 Discover PaperBanana

While writing your research paper, you find this friendly tool that turns simple text descriptions into professional diagrams and charts.

2
📦 Set it up on your computer

You download and install it easily, like adding a new app to help with your work.

3
🧠 Link a free AI helper

A quick guide walks you through connecting a no-cost AI service so it can understand and create visuals.

4
✨ Describe your research idea

You paste in your method explanation and tell it what the diagram should show, like 'overview of our framework'.

5
⚡ Watch it create and refine

The tool thinks step-by-step, plans the layout, styles it beautifully, and improves it a few times until perfect.

✅ Get publication-ready images

You receive stunning diagrams or charts saved as images, ready to drop into your paper and impress readers.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 36 to 1,083 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is paperbanana?

PaperBanana is a Python tool that turns text descriptions of research methods into publication-ready diagrams and statistical plots, solving the pain of manually drawing figures for papers. Feed it a methodology excerpt and caption via CLI (`paperbanana generate --input method.txt --caption "Framework overview"`), and it spits out polished PNGs using free Google Gemini APIs. It's an open-source implementation and extension of Google Research's PaperBanana paper, now handling slide generation and more academic visuals.

Why is it gaining traction?

It skips pricey tools or custom scripts—zero-cost setup with a free API key, plus CLI, Python API, and MCP server for seamless IDE integration like Claude Code or Cursor. Built-in evaluation compares your outputs to human references on faithfulness, readability, and aesthetics. Developers love the iterative refinement that delivers NeurIPS-style results without design skills.

Who should use this?

AI researchers and PhD students cranking out methodology diagrams for arXiv submissions. ML engineers prototyping vit implementation github or yolo implementation github projects needing quick architecture visuals. Academic teams extending llama implementation github or graphrag implementation github with research slides.

Verdict

Worth pip-installing for academic diagram needs—CLI works out of the box, docs are thorough, tests cover core flows. 259 stars and 1.0% credibility score mean it's alpha-stage; production users should monitor stability.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.