A high-performance OpenAI-compatible HTTP server that uses DuckDuckGo's AI backend, providing free access to multiple AI models through the familiar OpenAI API interface.
The easiest way to get started is using the pre-built Docker image:
# Pull the Docker image
docker pull amirkabiri/duckai
# Run the container
docker run -p 3000:3000 amirkabiri/duckaiThe server will be available at http://localhost:3000.
Docker image URL: https://hub.docker.com/r/amirkabiri/duckai/
- Clone the repository:
git clone git@github.com:amirkabiri/duckai.git
cd duckai- Install dependencies:
bun install- Start the server:
bun run devimport OpenAI from "openai";
const openai = new OpenAI({
baseURL: "http://localhost:3000/v1",
apiKey: "dummy-key", // Any string works
});
// Chat completion
const completion = await openai.chat.completions.create({
model: "gpt-4o-mini", // Default model
messages: [
{ role: "user", content: "Hello! How are you?" }
],
});
console.log(completion.choices[0].message.content);DuckAI OpenAI Server bridges the gap between DuckDuckGo's free AI chat service and the widely-adopted OpenAI API format. This allows you to:
- Use multiple AI models for free - Access GPT-4o-mini, Claude-3-Haiku, Llama-3.3-70B, and more
- Drop-in OpenAI replacement - Compatible with existing OpenAI client libraries
- Tool calling support - Full function calling capabilities
- Streaming responses - Real-time response streaming
- ✅ Rate limiting - Built-in intelligent rate limiting to respect DuckDuckGo's limits
gpt-4o-mini(Default)o3-miniclaude-3-haiku-20240307meta-llama/Llama-3.3-70B-Instruct-Turbomistralai/Mixtral-8x7B-Instruct-v0.1
- ✅ Chat completions
- ✅ Streaming responses
- ✅ Function/tool calling
- ✅ Multiple model support
- ✅ Rate limiting with intelligent backoff
- ✅ OpenAI-compatible error handling
- ✅ CORS support
- ✅ Health check endpoint
- Bun runtime (recommended) or Node.js 18+
- Clone the repository:
git clone git@github.com:amirkabiri/duckai.git
cd duckai- Install dependencies:
bun install- Start the server:
bun run devThe server will start on http://localhost:3000 by default.
import OpenAI from "openai";
const openai = new OpenAI({
baseURL: "http://localhost:3000/v1",
apiKey: "dummy-key", // Any string works
});
// Basic chat completion
const completion = await openai.chat.completions.create({
model: "gpt-4o-mini",
messages: [
{ role: "user", content: "Hello! How are you?" }
],
});
console.log(completion.choices[0].message.content);const tools = [
{
type: "function",
function: {
name: "calculate",
description: "Perform mathematical calculations",
parameters: {
type: "object",
properties: {
expression: {
type: "string",
description: "Mathematical expression to evaluate"
}
},
required: ["expression"]
}
}
}
];
const completion = await openai.chat.completions.create({
model: "gpt-4o-mini",
messages: [
{ role: "user", content: "What is 15 * 8?" }
],
tools: tools,
tool_choice: "auto"
});
// The AI will call the calculate function
console.log(completion.choices[0].message.tool_calls);const stream = await openai.chat.completions.create({
model: "gpt-4o-mini",
messages: [
{ role: "user", content: "Tell me a story" }
],
stream: true
});
for await (const chunk of stream) {
const content = chunk.choices[0]?.delta?.content;
if (content) {
process.stdout.write(content);
}
}curl -X POST http://localhost:3000/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer dummy-key" \
-d '{
"model": "gpt-4o-mini",
"messages": [
{"role": "user", "content": "Hello!"}
]
}'POST /v1/chat/completions- Chat completions (compatible with OpenAI)GET /v1/models- List available modelsGET /health- Health check endpoint
PORT- Server port (default: 3000)HOST- Server host (default: 0.0.0.0)
docker build -t duckai .docker run -p 3000:3000 duckai- Fork the repository
- Create a feature branch
- Make your changes
- Add tests for new functionality
- Run the test suite
- Submit a pull request
MIT License - see LICENSE file for details.
This project is not affiliated with DuckDuckGo or OpenAI. It's an unofficial bridge service for educational and development purposes. Please respect DuckDuckGo's terms of service and rate limits.