SpicyAPI developer documentation

One API.
Every multimodal model.

Image and video generation runs as tasks: submit one, get an ID immediately and receive the result by webhook. Text models answer through OpenAI-, Anthropic- and Gemini-compatible endpoints. Inputs follow each model's schema, and failed tasks are not billed.

Start here Quickstart Error reference

GET /console/v1/catalog/models

curl "https://api.spicyapi.ai/console/v1/catalog/models"

# No account or API key needed: every public model and its price.
# To call a model, take its ID from GET /api/v1/models with your key.

One API. Every multimodal model.

One API.
Every multimodal model.