Any2Poster
any2poster.github.io
- Windows
- macOS
- Linux
- Self-hosted
What Any2Poster is
Any2Poster is a free, open-source project that converts documents, slides, notebooks, and videos into academic posters. It accepts PDF, DOCX, PPTX, LaTeX, Markdown, plain text, Jupyter notebooks, HTML, web addresses, and YouTube URLs. Outputs include PDF, PNG, and HTML artifacts, with custom poster dimensions and print-ready export. Its six-stage workflow parses, chunks, analyzes, plans, generates, and compiles content; checkpoints let a run resume from cached stages. Poster text is rendered through HTML and CSS with Playwright, which the project says avoids typography hallucination. Suitable source figures can be preserved, while AI generation fills panels lacking an appropriate extracted figure. An optional vision-language-model review checks overflow, blank space, font size, figure balance, and reading order. Setup requires Python 3.10 or newer, Playwright Chromium, and an OPENROUTER_API_KEY. Installation and Docker instructions cover Windows, macOS, and Linux.
Who it suits
Any2Poster suits researchers, students, and technical users creating academic posters from varied source material. It is intended for people comfortable installing and running an open-source project.
Good to know, and what to watch
- Accepts documents, notebooks, webpages, and videos
- Exports PDF, PNG, and HTML
- Checkpointed workflow can resume cached stages
- Optional panel-level visual feedback
- Supports custom poster sizes
- Requires Python 3.10 or newer
- Requires Playwright Chromium installation
- Needs an OPENROUTER_API_KEY
- Web interface is listed as coming soon
The verdict
Any2Poster provides a broad, source-driven poster workflow with print-ready outputs and optional visual checks. Its open-source setup offers flexibility, but requires local installation and an API key.
What the maker publishes
- API key requirement
- The README says the only required key to get started is OPENROUTER_API_KEY. source
- Availability
- The GitHub repository describes Any2Poster as open source and links an MIT license. source
- Availability
- The repository describes Any2Poster as open source and displays an MIT license. source
- Checkpointing
- The six pipeline stages are checkpointed, and a run can resume from cached checkpoints. source
- Deployment
- The repository documents installation from source and provides Docker build and run instructions. source
- Evaluation
- Any2Poster Bench evaluates poster generation across input formats and domains using quiz scoring and vision-language-model judgments. source
- Feedback
- An optional panel-level vision-language model feedback loop checks for issues such as overflow, blank space, font size, figure balance, and reading order. source
- Figures and data
- The project says it preserves suitable figures extracted from the source and uses AI generation for panels without a suitable extracted figure. source
- Figures and data
- The project says it preserves suitable figures extracted from the source and validates chart values against source text before image generation. source
- Founded
- 2026 source
- Input formats
- It accepts PDF, DOCX, PPTX, LaTeX, Markdown, plain text, Jupyter notebooks, HTML or URLs, and YouTube URLs. source
- Input formats
- It accepts PDF, DOCX, PPTX, LaTeX, Markdown or plain text, Jupyter notebooks, HTML or URLs, and YouTube video URLs. source
- Installation requirement
- The README lists Python 3.10 or newer as a requirement and provides installation instructions for Windows, macOS, and Linux. source
- Integrations
- The pipeline uses Docling for parsing, Playwright for rendering, and lists Gemini 3 Pro for AI visual synthesis. source
- Integrations
- The documented setup requires an OpenRouter API key; the project also names Docling, Playwright, and Gemini 3 Pro in its pipeline. source
- Intended audience
- The project describes its output as academic posters and provides an evaluation benchmark for poster generation across input formats and content domains. source
- Output formats
- It can produce PDF and PNG posters, along with an HTML artifact. source
- Outputs
- The project lists PDF, PNG, and HTML artifact outputs. source
- Purpose
- Any2Poster turns documents, slideshows, notebooks, or videos into publication-quality academic posters. source
- Purpose
- Any2Poster turns documents, slideshows, notebooks, and videos into publication-quality academic posters. source
- Quality feedback
- An optional panel-level vision-language-model feedback loop checks issues such as overflow, blank space, font size, figure balance, and reading order. source
- Rendering
- Poster text is rendered with HTML/CSS via Playwright, and the README says this avoids typography hallucination. source
- Rendering
- Poster text is rendered with HTML and CSS using Playwright, which the project says avoids typography hallucinations. source
- Requirements
- The project lists Python 3.10 or later and requires installing the Playwright Chromium browser for rendering. source
Checked against the maker's pages on 8 Oct 2026.