Start with the quickest path
Use Quick Start when you want the shortest route to an image generation request, then open API Reference for exact request bodies, response shapes, headers, and errors.
One API key for image and video generation models. Compatible with OpenAI-style APIs, provider routing, and credit-aware billing.
Pixapi keeps provider details behind a stable API surface. Your team can ship against one request format while managing model vendors, usage tracking, and billing policy from the platform layer.
The Pixapi API docs are written for teams that need to move from a first request to a production workflow quickly. The guide is organized around real product tasks: image generation, image editing, video generation, async tasks, billing, and model selection all connect back to the same account and credit system.
Use Quick Start when you want the shortest route to an image generation request, then open API Reference for exact request bodies, response shapes, headers, and errors.
Before coding, check which endpoints are synchronous, which are async-only, which field carries the model id, and how generated assets are returned.
Use the Model Catalog to compare Gemini image models, Nano Banana aliases, GPT Image models, Veo, Wan, and other supported routes.
Read the Async Tasks guide when a workflow may take longer than a normal request. Video generation and large media jobs should be submitted as tasks, then polled through the task endpoint.
Review Billing and Credits before opening a feature to users. Credits connect model cost to outputs, quality, resolution, duration, and provider-specific settings.
Pixapi exposes llms.txt and llms-full.txt so coding agents can follow current endpoint names, model ids, and async patterns.
Use these pages to choose a model, price the workflow, write the request, handle the response, monitor the task, and support the customer. When one prototype grows into several visual AI features, the docs keep integrations aligned under one API key, one credit ledger, and one developer workflow.