Skip to main content
Fhddos provides a complete OpenAI-compatible proxy layer that lets you call OpenAI, Gemini, Claude, Doubao, Kling, and other AI providers through a single base URL and token. This page introduces the core concepts — unified authentication, multi-cloud routing, and compatibility guarantees — and shows you how to make your first request in under a minute.
Set Authorization: Bearer <fhddos-token> on every request. Your Fhddos token starts with oh-.

​
How it works

Fhddos sits between your application and the upstream AI providers. Because the API surface mirrors OpenAI’s REST specification, you only need to change two things in your existing code:
  • base_url → https://aiapi.fhddos.com
  • api_key → your Fhddos token (starts with oh-)
Your prompts, message arrays, tool definitions, and streaming logic stay exactly the same.
If your account connects to both official and economy provider groups, restrict your OpenAI compatibility promise to the official group. See the video-generation page for details on group isolation.

​
Set up your environment

Export these two variables once and every subsequent curl or SDK example on this site will work without modification. 
export BASE_URL="https://aiapi.fhddos.com"
export TOKEN="oh-xxxxxxxxxxxxxxxx"

​
Quick test

Verify your token is valid and see which models you have access to: 
curl "$BASE_URL/v1/models" \
  -H "Authorization: Bearer $TOKEN"
A successful response returns a JSON list object whose data array contains model objects. If you see an 401 error, double-check that your token begins with oh- and is correctly set in the Authorization header.

​
Key capabilities

FeatureDescription
Official compatibilityUnder official groups and channels, the API stays aligned with OpenAI’s REST spec. Reuse existing SDKs and code with zero logic changes.
Unified authenticationOne Fhddos token accesses multiple providers. Apply fine-grained key permissions and usage monitoring from the Fhddos console.
Multi-cloud routingSwitch backend providers or enable redundancy strategies from the console without touching your application code.

​
Explore the API

Chat Completions

Classic conversational interface with streaming, tool calling, and JSON mode.

Responses

Unified multimodal interface supporting text, images, and structured output in one request.

Audio

Text-to-speech, audio transcription, and translation with Whisper models.

Images

Generate and edit images with gpt-image-2, including async task workflows.

Video

Create Sora videos, poll for completion, and download MP4 results.

Models

List all models your token is authorized to use.