PosterGen

PosterGen is a self-hosted tool that turns research-paper PDFs into conference posters using a team of AI agents. It suits researchers who want a structured route from paper to both a poster image and an editable PowerPoint, and are comfortable configuring software and model APIs. Its academic focus and editable output are compelling; setup requirements and required logos make it less suitable for anyone seeking a ready-to-use design app.

Overview

PosterGen breaks poster creation into stages: one agent extracts and structures paper content, another plans the narrative and visual placement, and layout and styling agents arrange and dress the result. The documented workflow uses a three-column layout, with a balancing sub-agent working to use columns well and prevent overflow. This division gives the process a clear academic-poster logic rather than treating a paper as raw material for a generic graphic.

The tool produces a PNG poster and an editable PowerPoint file. It draws color choices from an affiliation logo and applies professional typography and keyword highlighting. That combination is useful when a poster must be shared as an image but also revised afterward. The software is MIT-licensed and acknowledges LangGraph, Marker and python-pptx as underlying open-source projects; the license provides the software as-is, without warranty.

Key features

Paper-to-poster workflow

PosterGen parses a research PDF, plans a storyboard, then calculates placement and spacing. The approach is tailored to turning paper content into a conference-poster narrative. It is not simply a blank-canvas design tool: users seeking broad-purpose graphics or a hands-off hosted service may prefer another option.

Configuration and outputs

A local web interface accepts the PDF and logos, lets users configure models and dimensions, and provides generated files for download. A configuration file allows changes to layout, typography, color generation, visual-asset sizing and content optimization. Custom dimensions are supported, within a documented width-to-height ratio range of 1.4 to 2. The output directory also records runtime and API cost in a JSON log, which can help users track their generation runs.

Pricing

PosterGen is free, with a free plan. There is no paid tier described. Free software does not mean a zero-effort setup: users need configured API keys, and model API usage can incur costs. The project logs API cost alongside runtime, a useful consideration for repeated generation.

Platforms

The project supports Windows, Linux and macOS and provides a local web interface; it is also categorized as self-hosted. Setup requires Python 3.11 and LibreOffice. The web interface additionally requires Node.js and configured API keys. Users must supply a research-paper PDF, an affiliation logo for color extraction and a conference logo. These requirements favor users who can manage a local software environment over those who want to start designing immediately in a browser.

Documented API-key providers include OpenAI, Anthropic and Zhipu, and configurable base URLs cover OpenAI-, Anthropic-, Google- and Zhipu-compatible services. Documented text and vision model options include GPT-4.1, GPT-4o, GPT-4.1 mini and Claude Sonnet 4.

Who it's for

PosterGen is best suited to researchers preparing conference posters from existing papers, especially those who value a repeatable workflow and an editable PowerPoint deliverable. It is a weaker fit for users without the required files, API configuration or willingness to install its dependencies, and for anyone seeking a general-purpose poster editor.

Pros and cons

Pros

• Research-focused workflow: Parsing, storyboarding and layout are separated into stages built around turning a paper into a poster narrative.
• Two useful deliverables: PNG and editable PowerPoint outputs support both sharing and later revision.
• Meaningful customization: Configuration covers layout, typography, colors, visual-asset sizing, content optimization and dimensions.
• Free and MIT-licensed: There is no stated purchase price, though API usage may still cost money.

Cons

• Setup takes technical effort: Python 3.11, LibreOffice, Node.js for the interface and configured API keys add friction before the first poster.
• Inputs are specific: Each workflow requires a paper PDF and both affiliation and conference logos.
• API usage is a separate cost consideration: Free software does not remove potential model charges.
• Scope is narrow: Its design workflow targets academic conference posters, not general poster-making needs.

Alternatives

For a broader selection of tools in the category, browse AI Poster Generators.

• Ridvay is a better fit if you want mobile apps alongside web access and a free allowance that includes daily AI designs and images, plus manual designs and templates.
• Bibby AI is worth considering for institutional needs; its Institution plan is custom to volume licensing and includes SSO and SAML, API access, an on-premise option and an SLA.
• Kittl is a web-based alternative with a free tier capped at five projects and a one-time allotment of 100 AI tokens.
• PosterMont offers a free guest quota for poster images and short videos without an account or card; guest images carry a small watermark and are not saved.
• Piktochart may suit teams that want a web-based tool with a free allowance of two downloads, 1 GB storage and 60 AI credits.
• Pixazo AI Research Poster Maker is a browser-based option, though image generation uses a Pixazo Playground account.
• Any2Poster is another free option for Linux, macOS and Windows.
• LightX AI Poster Generator has mobile and web access and a free Starter tier with six daily AI credits.

Verdict

Choose PosterGen if you are preparing a research conference poster and want a paper-centered workflow that delivers both a PNG and editable PowerPoint without a software purchase price. Its main advantage is the combination of academic structure, configurable design and editable output. Look elsewhere if you need an immediate, low-setup design experience or do not want to supply logos, manage dependencies and configure model APIs.