Cloudflare AI Web

中文 ｜ English

Deployment

Vercel

Example: https://ai.jaze.top

It may not respond when the quota is used up, it is recommended to deploy it yourself

Docker

docker run -d --name cloudflare-ai-web \
  -e CF_ACCOUNT_ID=YOUR_CF_ACCOUNT_ID \
  -e CF_WORKERS_AI_TOKEN=YOUR_CF_WORKERS_AI_TOKEN \
  -p 3000:3000 \
  --restart=always \
  jazee6/cloudflare-ai-web

Features

• Quickly build a multimodel AI platform using Cloudflare Workers AI
• Support Cloudflare AI Gateway to access models such as Gemini
• Support fast deployment with Serverless
• Chat history is stored locally
• Image attachments are resized in the browser (long edge up to 1568px, at most 512 KiB each, up to 5 per request) so requests fit the Vercel Functions body limit
• Access Session protection via deployment password

Note: In public mode anyone can use your inference APIs. Set APP_PASSWORD to enable Access Session.

Deployment Instructions

Environment Variables

Name	Description	Required
CF_ACCOUNT_ID	Cloudflare Account ID	✅
CF_WORKERS_AI_TOKEN	Cloudflare Workers AI Token	✅
CF_AI_GATEWAY_NAME	Cloudflare AI Gateway Name
CF_AI_GATEWAY_TOKEN	Cloudflare AI Gateway Auth Token	With gateway
GOOGLE_API_KEY	Google AI Studio Token	With Google
CF_AI_GATEWAY_PROVIDERS	Cloudflare AI Gateway Providers
APP_PASSWORD	Access Password (Access Session)

CF_WORKERS_AI_TOKEN

• Manage Account - Account API Tokens - Create Token - Create with Workers AI template

CF_AI_GATEWAY_PROVIDERS

Supported providers:

• google

Multiple providers are separated by commas

Sponsor

Click Me