Deploy Free DeepSeek API Gateway on Cloudflare Workers Serverless

Project Specifications & Benefits

• $0 Server & AI Costs: Runs entirely on Cloudflare Workers AI free tier (10,000 queries per day).
• OpenAI-Compatible: Translates parameters on the fly so it integrates natively with Cursor, VS Code, and Obsidian.
• Secure Bearer Auth: Prevents unauthorized third parties from consuming your free Cloudflare limits.
• Edge Runtime Speed: Resolves calls directly at Cloudflare's closest regional edge nodes (under 50ms wrapper overhead).

How It Works: OpenAI-to-Cloudflare Gateway

Standard development tools like Cursor expect chat API responses in the official OpenAI format (containing specific keys like choices, message, and content). However, Cloudflare Workers AI uses its own native response structure. Our proxy intercepts the request, validates the custom Authorization header, queries the Cloudflare AI library, and wraps the output in an OpenAI-compliant JSON object:

Edge Computing: DeepSeek-R1 inference processed on Cloudflare's serverless nodes.

The API Gateway Code (Index.js)

If you choose not to use the 1-click installer, you can write the code manually. Below is the full JavaScript file (index.js) that processes CORS requests, verifies your custom key, and runs the DeepSeek model:

export default {

 async fetch(request, env) {

 if (request.method === "OPTIONS") {

 return new Response(null, {

 headers: {

 "Access-Control-Allow-Origin": "*",

 "Access-Control-Allow-Methods": "GET, POST, OPTIONS",

 "Access-Control-Allow-Headers": "Content-Type, Authorization",

 },

 });

 }

 if (request.method !== "POST") return new Response("POST only", { status: 405 });



 const authHeader = request.headers.get("Authorization");

 if (!authHeader || authHeader !== `Bearer ${env.API_KEY}`) {

 return new Response(JSON.stringify({ error: "Unauthorized" }), {

 status: 401,

 headers: { "Content-Type": "application/json", "Access-Control-Allow-Origin": "*" },

 });

 }



 try {

 const body = await request.json();

 const messages = body.messages || [];

 const model = body.model || "@cf/deepseek-ai/deepseek-r1-distill-qwen-32b";



 const aiResponse = await env.AI.run(model, { messages });



 return new Response(JSON.stringify({

 id: `chatcmpl-${crypto.randomUUID()}`,

 object: "chat.completion",

 created: Math.floor(Date.now() / 1000),

 model: model,

 choices: [{

 index: 0,

 message: { role: "assistant", content: aiResponse.response || aiResponse.text || "" },

 finish_reason: "stop"

 }],

 usage: { prompt_tokens: 0, completion_tokens: 0, total_tokens: 0 }

 }), {

 status: 200,

 headers: { "Content-Type": "application/json", "Access-Control-Allow-Origin": "*" }

 });

 } catch (err) {

 return new Response(JSON.stringify({ error: err.toString() }), {

 status: 500,

 headers: { "Content-Type": "application/json", "Access-Control-Allow-Origin": "*" }

 });

 }

 }

};

Manual Setup & Deployment Steps

If you want to configure wrangler and write files step-by-step, follow these instructions in your local project folder:

Step 1: Setup wrangler.json

Create a wrangler.json file in your directory to bind the AI catalog to your serverless instance:

{

 "name": "free-deepseek-api",

 "main": "index.js",

 "compatibility_date": "2026-07-09",

 "ai": {

 "binding": "AI"

 }

}

Step 2: Bind API Key & Deploy

Run these terminal commands to secure your endpoint, log in to Cloudflare, and push your worker code to the edge network:

1. Run npx wrangler login in your terminal.
2. Save your secure key: echo your-api-key | npx wrangler secret put API_KEY
3. Run npx wrangler deploy to live stream your worker to the cloud.

Terminal Output: Successful deploy showing the workers.dev endpoint URL.

Integrating with Cursor & VS Code (Continue.dev)

Once deployed, you can paste your worker details into your IDE to bypass paid limits and enjoy unlimited AI reasoning.

For Cursor:

1. Go to Settings > Models > OpenAI-Compatible in Cursor.
2. Set the Base URL to your deployed worker link: https://free-deepseek-api.YOUR_SUBDOMAIN.workers.dev/v1
3. Enter the custom API Key you created.
4. Set the model name to: @cf/deepseek-ai/deepseek-r1-distill-qwen-32b (or @cf/qwen/qwen1.5-14b-chat-awq).

Configuration Mockup: Entering the custom endpoint and model details inside Cursor settings.

Frequently Asked Questions (FAQs)