API documentation hub
Learn what APIQIK offers, follow the onboarding steps, and browse OpenAI-compatible gateway endpoints.
About APIQIK
APIQIK is an AI gateway for teams and developers: one Base URL and one auth style to reach chat, image, and other hosted models, while keeping OpenAI-compatible request shapes.
You focus on prompts and model choice; routing, billing, and entitlements are governed by the platform and your token policies—less bespoke multi-vendor wiring.
Broad model coverage
Chat, image, and more—availability and pricing follow the model marketplace and console.
Reliable streaming
Streaming and non-streaming responses for chat, tools, and long-form workflows.
Security-first
HTTPS, Bearer auth, per-token quotas and model scope—never ship secrets to the browser.
Easy migration
Keep the OpenAI SDK; swap base_url and api_key to send the first request.
Core capabilities
Chat completions
Customer support, writing, coding assistants; multi-turn messages and streaming.
- Request bodies align with OpenAI Chat Completions
- Tune parameters such as temperature when the model supports them
- Check marketplace notes for context limits and billing multipliers
Image generation
Call hosted image models through the OpenAI-compatible Images path.
- Mind model, prompt, size, and related fields
- URLs vs base64 depend on the model
- Billing is reflected in console and logs
Video and audio
When enabled in your workspace, use the compatible routes exposed by the gateway (subject to availability).
- Confirm model names and entitlements in the console
- Long jobs may be async—follow model docs
- See per-model guidance for parameters
Catalog and usage
List models available to a token and inspect usage in the console.
- Use GET /v1/models before integration
- Trace spend and payloads via console logs
- 401/402/404/429—check auth, quota, names, and backoff