#5151st of 69 in Data and scraping for AIFor science, health and research, science, research and papers
paper2poster
Turns a scientific paper PDF into an editable PowerPoint poster
Live demo ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 4.1k
- License
- MIT
- Last commit
- Nov 2026
- Language
- Python
3,588 on 25 Apr 20264,052 on 14 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
PosterAgent takes a paper.pdf and produces an editable poster.pptx through a parser, a binary-tree layout planner, and a painter-commentor loop that uses VLM feedback to fix overflow and alignment. It runs with API models such as GPT-4o or with local Qwen-2.5 models served by vLLM. The repo also includes the Paper2Poster benchmark and evaluation scripts (PaperQuiz, VLM-as-Judge, statistical metrics).
Who it is for: Researchers who want a first-draft conference poster from a paper PDF
Strengths
- Outputs editable .pptx rather than a flat image
- Mixes LLM and VLM backends: GPT-4o, or Qwen-2.5 via vLLM
- Ships evaluation suite: PaperQuiz, VLM-as-Judge, statistical metrics
- Auto-adds conference logos; YAML style config; parallel section generation
Weaknesses
- Needs LibreOffice and poppler installed alongside Python requirements
- Local vLLM models require editing the port in get_agent_config()
- CLI scripts only; the README documents no packaged API or web UI
- No tagged releases
What it needs
- GPU optional
- Docker
- Needs LibreOffice, poppler, OpenAI API key (for GPT-4o), vLLM (for local models), Google Custom Search API (optional)
- Models: GPT-4o, Qwen-2.5-7B-Instruct (vLLM), Qwen VL (vLLM)
Also in Data and scraping for AI
See all 69| Rank | Project | Score |
|---|---|---|
| 1 | mineruParses PDFs, images and Office files into Markdown and structured JSON | 91 out of 100 |
| 2 | doclingPython library that converts PDFs, Office files and more into structured text | 91 out of 100 |
| 3 | opendataloader-pdfPDF parser that outputs Markdown, JSON with bounding boxes, and Tagged PDF | 91 out of 100 |
| #4 | agent-reachCLI that gives coding agents read access to login-gated and Chinese web platforms | 88 out of 100 |
| #5 | markitdownPython library and CLI that converts office files, PDFs and audio to Markdown | 87 out of 100 |