---
title: "Deploy Banana Slides"
description: "AI presentation generator: prompt, outline or doc to slides (Gemini/OpenAI)"
category: "AI/ML"
url: https://railway.com/deploy/banana-slides-1
---

# Deploy Banana Slides

AI presentation generator: prompt, outline or doc to slides (Gemini/OpenAI)

**[Deploy Banana Slides on Railway](https://railway.com/template/banana-slides-1)**

Machine-readable deploy manifest (JSON, validated by TemplateCI): https://railway.com/deploy/banana-slides-1/manifest.json

- **Creator:** Agaz Self-Host
- **Category:** AI/ML
- **Total deploys:** 10

## Template content

### Banana Slides https://raw.githubusercontent.com/Anionex/banana-slides/v0.4.0/frontend/public/logo.png

- **Source:** https://github.com/aalfath/banana-slides-railway-template
- **Health check:** /health
- **Public domain:** Yes

### Banana Slides Backend https://raw.githubusercontent.com/Anionex/banana-slides/v0.4.0/frontend/public/logo.png

- **Source:** https://github.com/aalfath/banana-slides-railway-template
- **Health check:** /health

## Documentation

# Deploy and Host Banana Slides on Railway

Banana Slides is an open-source (AGPL-3.0) AI presentation generator. You give it a one-line idea, an outline or page descriptions, optionally with reference documents and a style template image, and it writes the outline, generates every slide as an image with Gemini or an OpenAI-compatible model, lets you edit pages by prompt, and exports PPTX or PDF.

## About Hosting Banana Slides

This template runs Banana Slides v0.4.0 as two services built from this repo, with the official upstream images pinned by digest. The backend (Flask) keeps its SQLite database and every upload and generated slide on a Railway volume, and runs the database migrations on each start. The public web service is nginx: it serves the React app, proxies the API over Railway's private network, and puts the whole UI behind a login with a generated password, because Banana Slides has no accounts of its own and its Settings page holds your AI API key. After deploy, open the URL, sign in, and add your API key under Settings.

## Common Use Cases

* Turn a one-line idea or an outline into a complete slide deck
* Re-style a document or report as slides that follow your own template image
* Edit individual slides by prompt and export the result as PPTX or PDF

## Dependencies for Banana Slides Hosting

* An API key for Google Gemini (default) or for any OpenAI-compatible endpoint that serves text and image models
* A Railway volume for the SQLite database, uploads and generated slides (included)

### Deployment Dependencies

* Banana Slides source: https://github.com/Anionex/banana-slides
* Template repo: https://github.com/aalfath/banana-slides-railway-template
* Google AI Studio API keys: https://aistudio.google.com/apikey

### Implementation Details

Services in this template:

* **Banana Slides** (public): nginx on `PORT` 8080. It serves the UI, proxies `/api` and `/files` to the backend, and requires the login `admin` with the password in its `BASIC_AUTH_PASSWORD` variable (generated at deploy; read it in the service's Variables tab). `/health` answers without login for Railway's health check.
* **Banana Slides Backend** (private): the Flask API on port 5000, listening on IPv6 for Railway's private network. Its volume at `/data` holds `instance/database.db` and `uploads/`. `SECRET_KEY` is generated at deploy.

After deploy:

1. Open the public URL and sign in as `admin` with the generated `BASIC_AUTH_PASSWORD`.
2. Open Settings, choose the provider format (Gemini or OpenAI), enter your API key (and API base URL if you use a proxy or another OpenAI-compatible service), and pick the text and image models. Settings are stored in the database on the volume, so they survive redeploys.
3. Optional: set the output language in Settings (the upstream default is Chinese), and add a MinerU token for parsing PDFs and Office files, or a Baidu OCR key for editable PPTX export.

You can also set `GOOGLE_API_KEY`, or `AI_PROVIDER_FORMAT=openai` with `OPENAI_API_KEY` and `OPENAI_API_BASE`, as variables on the backend service. They seed the settings on the first start only.

Updates: this template builds from the GitHub repo above, so when the pinned version is bumped, Railway shows an "update available" notice on your project and you choose when to apply it. Read `CHANGELOG.md` in the repo first.

## Why Deploy Banana Slides on Railway?


Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it.

By deploying Banana Slides on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway.



## Similar templates

- [Chat Chat](https://railway.com/deploy/-WWW5r) — Chat Chat, your own unified chat and search to AI platform.
- [Weaviate](https://railway.com/deploy/weaviate-1) — Vector database with hybrid search and an API key-protected API
- [Möbius Personal AI OS](https://railway.com/deploy/mobius-v01-personal-ai-os) — Self-hosted AI agent that builds and runs apps for you, with memory.

Open this page in a browser: https://railway.com/deploy/banana-slides-1
