HiAPI

HiAPI is a production-focused AI API gateway that helps you generate images, video, and audio through a single, consistent interface. Submit a job, track status, and retrieve results as persistent artifact links—so you don’t need to build custom storage for each provider.

Use the same task lifecycle across models: pass the model name to target the right image/video/audio generator, provide model-specific parameters, and optionally enable persistent outputs. You can poll for completion or use callbacks to get notified when a job finishes.

HiAPI also supports agent-friendly integration paths with MCP, Skills, and an llms.txt prompt so AI agents can discover models and call the API using standardized workflows. Each model includes visible pricing details and parameters, making it easier to size usage before you ship.

What can I generate with HiAPI?

HiAPI supports AI media generation including images, video, and audio (and it also includes text-generation endpoints in its unified interface). Choose a model name and submit a task with the model’s required input parameters.

How does the HiAPI async task API work?

You create a job with a POST to the /v1/tasks endpoint, receive a task id, and then either poll for completion or use a callback. Once finished, you retrieve durable artifact links for the generated outputs.

What are persistent artifact links?

Persistent artifacts return durable links to the generated outputs, so your workflow can pass results onward without managing provider-specific storage yourself. Enable persistence via the task request.

Can I use callbacks instead of polling?

Yes. When you submit a task you can provide a callback URL and completion timing, so you’ll get notified when generation is done rather than repeatedly checking task status.

How do I route a request (default vs beta/ext)?

You can set an optional route field such as a beta or ext route. Use routing to select an available path for cost comparison and execution behavior while keeping the same overall task API flow.

Do model prices show up before I integrate?

Yes—each model lists pricing details and parameters. Review model-specific cost before launching to avoid surprises when you start sending real generation jobs.