Skip to main content
The Thrads Supply-Side Platform (SSP) API enables publishers to run real-time first-price auctions across multiple DSPs (Demand-Side Platforms) and receive winning ad creatives in a single request.

Visit the Thrads Platform

Manage your chatbots, API keys, and analytics

​
Get Your API Key

  1. Go to the Thrads Platform and sign up as a publisher.
Access is currently limited to approved partners. After signing up, your application will be reviewed within 24 hours. Once approved, you can create your chatbot instance and generate API keys from the platform dashboard.
  1. Once approved, create a chatbot in the platform and you’ll be provided a staging API key to test your integration.
  2. When you’re ready to go live, contact us at contact@thrads.ai and you’ll be provided a production API key.

​
Authentication

All API requests require authentication using an API key in the request header: 
thrad-api-key: your-api-key
For browser-based requests, you’ll also need to register your domain for CORS support. Contact us at contact@thrad.ai to register your domains.

​
Staging vs. Production

Use your staging API key during development and your production API key in your live environment.
  • Staging key:
    • Ads are returned for testing purposes but are not counted for monetization
    • An ad will be served on every request to facilitate testing and integration
  • Production key:
    • Ads are live and contribute to monetization
    • You may receive "bid": null responses when no winning bid is found

​
Implementation Flow

1

User sends a message

Capture the user message and add it to your conversation history.
2

Call Thrad Bid Request API

Send the full conversation history to the Thrad bid request endpoint. Always include all messages from the session — the API uses the full context to select the most relevant ad.
3

Handle Response

  • If bid is returned: Render the ad creative
  • If no bid: Continue without showing an ad
  • If error: Log for debugging and continue without ad
4

Render Ad Creative

Display the ad in your UI. See the Rendering Ads guide for templates and best practices.
Send the bid request before the assistant response is generated. The last message in the conversation should be the latest user message — there is no need to wait for your LLM to respond before calling the API. This lets the bid request run in parallel with your chatbot’s response, dramatically reducing perceived latency for the end user.

​
Rate Limits

The API supports up to 40 requests per second per API key. Rate limit information is included in response headers:
  • X-RateLimit-Limit: Maximum requests per second
  • X-RateLimit-Remaining: Requests remaining in current window
  • X-RateLimit-Reset: Unix timestamp when window resets

​
Response Format

All API responses follow a standardized format: 
{
  "requestId": "api_req_123",
  "timestamp": "2025-11-24T21:51:52.240297Z",
  "totalTime": 0.123,
  "status": "success",
  "message": "Bid successful",
  "data": { ... },
  "error": null
}

​
Guides

​
Support

For questions or support, contact us at contact@thrad.ai