Skip to content

Use Duckduckgo AI Chat to provide an OpenAI-compatible API that can be used for free with gpt-4o-mini.

License

Notifications You must be signed in to change notification settings

mumu-lhl/duckduckgo-ai-chat-service

Repository files navigation

Duckduckgo AI Chat Service

Docker Pulls GitHub Tag

English | 中文

Provide an OpenAI-compatible API for Duckduckgo AI Chat that can be used for free with gpt-4o-mini, claude-3-haiku, llama3.1...

Deploy

Render

Deploy to Render

Docker

docker run -d mumulhl/duckduckgo-ai-chat-service

Deno Deploy

Web

Fork this project, then visit https://dash.deno.com and create new project after loging in.

Command line

git clone https://github.com/mumu-lhl/duckduckgo-ai-chat-service --depth 1
deno install -A jsr:@deno/deployctl
deployctl deploy

Deno

# Run directly
deno run --allow-env --allow-net --unstable-cron https://raw.githubusercontent.com/mumu-lhl/duckduckgo-ai-chat-service/main/main.ts

# Run after installation
deno install -g --allow-env --allow-net --unstable-cron -n duckduckgo-ai-chat-service https://raw.githubusercontent.com/mumu-lhl/duckduckgo-ai-chat-service/main/main.ts
duckduckgo-ai-chat-service

Configuration

Configuration using environment variables:

  • TOKEN - Limit the tokens that can access the API, if you don't fill in, any token can access the API.
  • LIMIT - limit the request rate per second, default is 2
  • CLEAN_CACHE_CRON - how many hours to clean up the cache, default is 1

Usage

Just change the base url of the place where you need to use the OpenAI API to the one you deployed.