Introduction
A unified gateway for supported AI models, protocols, and task types.
The platform provides one account and API key system for models from many manufacturers. It standardizes authentication, model discovery, billing, and request logs while keeping internal provider routing and failover behind the gateway.
Start here
Quickstart
Send your first model request in a few minutes.
API Keys
Create a key and configure practical limits.
Models and Capabilities
Understand model IDs and choose a model.
API Reference
Browse every public endpoint, request field, and response shape.
Understand the integration
API Basics
Learn the shared endpoint and request conventions.
Protocols and SDKs
Choose the client protocol your application speaks.
Request Modes
Handle synchronous, streaming, and asynchronous work.
Call models by task
Text Generation and Chat
Build text, conversation, reasoning, and tool workflows.
Embeddings
Create vectors for semantic search and retrieval.
Image Generation
Generate and transform images.
Video Generation
Create and track asynchronous video tasks.
Additional task guides appear when their public APIs become available. Browse the Models page for the complete current catalog.
Operate your account
Use the Playground to test models, Request Logs and Usage to investigate calls, and Pricing and Billing to understand final charges.
Dynamic model information
Available manufacturers, models, capabilities, and prices can change. Use the Models page and each model's details as the current source of truth instead of copying a static list from the documentation.
Help us improve this page
Found something unclear, outdated, or incorrect?
Last updated on