https://github.com/vanna-ai/vanna-gpt-starter
Everything you need to run Vanna AI from ChatGPT
https://github.com/vanna-ai/vanna-gpt-starter
Last synced: 4 months ago
JSON representation
Everything you need to run Vanna AI from ChatGPT
- Host: GitHub
- URL: https://github.com/vanna-ai/vanna-gpt-starter
- Owner: vanna-ai
- Created: 2025-06-26T21:20:56.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2025-06-26T22:16:19.000Z (4 months ago)
- Last Synced: 2025-06-26T23:43:46.585Z (4 months ago)
- Language: Python
- Size: 1.95 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Vanna Custom GPT – Self-Hosted Starter Kit 🚀
This starter kit lets you run Vanna AI through your own Custom GPT by deploying a lightweight backend that forwards requests to the Vanna API.
## 🎥 Demo: Vanna Custom GPT in Action
[](https://www.youtube.com/watch?v=99SeFvRPs98)
---
## 🌟 What’s Inside
- `main.py` — Flask app to relay requests to Vanna
- `requirements.txt` — Python dependencies
- `Procfile` — For Railway/Heroku deployment
- `openapi.yaml` — Upload to your Custom GPT
- `.env.example` — Template for environment variables
- `README.md` — You’re reading it 🙂
---
## 🛠 Step 1: Set Your API Key
Create a `.env` file by copying the example:
```bash
cp .env.example .env
```
Then edit `.env` and add your Vanna API key:
```env
VANNA_API_KEY=your-api-key-here
```
---
## 🚀 Step 2: Deploy the Backend
### Option A: Google Cloud Run
```bash
gcloud auth login
gcloud run deploy vanna-api \
--source . \
--entry-point app \
--runtime python311 \
--port 8080 \
--allow-unauthenticated \
--region us-central1 \
--set-env-vars VANNA_API_KEY=your-api-key-here
```
---
### Option B: Railway (recommended for ease)
1. Go to [https://railway.app](https://railway.app) and log in
2. Create a new project and deploy this repo
3. Set an environment variable: `VANNA_API_KEY=your-api-key-here`
4. Railway will auto-detect the Flask app from `Procfile`
---
### Option C: Heroku
```bash
heroku login
heroku create vanna-api
heroku config:set VANNA_API_KEY=your-api-key-here
git push heroku main
```
---
## 🤖 Step 3: Set Up Your Custom GPT
1. Go to [https://chat.openai.com/gpts/new](https://chat.openai.com/gpts/new)
2. In the **Actions** tab, upload the `openapi.yaml` file
3. Update the `servers:` section in `openapi.yaml` with your deployed endpoint, e.g.:
```yaml
servers:
- url: https://your-cloudrun-or-railway-url-here
```
4. Save and start chatting!
---
## 🙋 Need Help?
Reach out to your Vanna contact or support@vanna.ai for assistance.