HiAPI

Core Features & Characteristics

Unified Async Task API: All image, video, and audio models have been upgraded to a unified asynchronous task API. Users submit tasks via the POST /v1/tasks endpoint (requiring model and input parameters, with optional storage and callback parameters) and retrieve results using the returned taskId. Supports calling every model through the same endpoint and schema.

Persistent Storage: Set storage to persistent, and generated outputs (artifacts) are stored and returned as durable links. No storage infrastructure to build or maintain, and artifact links stay reachable.

Callback Mechanism: Supports providing a callback URL (callback). HiAPI notifies you when the task reaches its final state, allowing you to skip polling.

AI Agent Friendly: Supports MCP (Agent-accessible tools), Skills (reusable agent instructions), and llms.txt, making it ready for any agent like Claude, Cursor, and Codex to discover, call, and compose HiAPI.

Production-Grade Reliability: Offers availability-first routing (designed for stable access, not just the cheapest route), optimization for long-running jobs with task tracking and callbacks, and 24/7 support.

Featured Supported Models

• GPT Image 2: OpenAI image generation and editing with sharp text rendering. Starts at $0.03/image for 1K resolution.
• Nano Banana 2: Google Gemini Flash image model, fast with strong character consistency. $0.05/image.
• Seedance 2.0: ByteDance cinematic video with native audio and multi-shot consistency.
• HappyHorse 1.1: Latest video model supporting text, image, and reference to video.
• FLUX 1.1 Pro: Black Forest Labs high-fidelity image generation, fast and detailed. $0.05/image.
• Qwen Image 2.0: Alibaba Qwen image model for high-quality generation and editing.

Target Users & Use Cases

• AI video SaaS teams: Quickly test video models, keep generated assets reachable, and skip building a storage path for every provider.
• Creative automation teams: Use persistent artifact links to easily pass outputs between campaign tools, review, and downstream automation.
• Agent workflow teams: Use a task API with callbacks to give agents a clean way to request, wait, retrieve, and reuse generated media.

Pricing & Cost Information

Pay-per-task model with flat rates. No token billing, no surge pricing. Every model lists its price up front for predictable costing:

• gpt-image-2-beta: $0.02/task
• Nano Banana 2 / flux-1.1-pro: $0.05/task
• 1080p Seedance clip: up to $0.823/task
• GPT Image 2: 1K resolution at $0.03/image, 2K at ~$0.04/image, and 4K at $0.06/image Volume discounts are available for larger usage (contact sales). New users get $1 free credit (approx. 50 images) upon signup to try real generations.

Frequently Asked Questions (FAQ)

• Which models does HiAPI support? HiAPI supports selected image, video, and audio generation models. Browse the full list on the models page.
• How do persistent artifact links work? Set storage to persistent, and generated outputs are stored for you and returned as durable links in the task result.
• Can I use callbacks instead of polling? Yes. Provide a callback URL and HiAPI notifies you when the task reaches its final state, so you skip polling.
• Can I try models before integrating? Yes. Use the Playground on each model page to explore parameters and example outputs before writing code.