The current published public catalog contains one visible model: nano-banana. In API requests, use the backend model IDDocumentation Index
Fetch the complete documentation index at: https://docs.eggapi.ai/docs/llms.txt
Use this file to discover all available pages before exploring further.
nanobanana.
| Public catalog name | API model value | Type | Task types | Base unit cost |
|---|---|---|---|---|
nano-banana | nanobanana | Image | image_generate, image_edit | $0.0100 per output image |
The backend code still contains older fallback configuration for additional internal models, but they are not public in the current published catalog and are intentionally not documented as public API options.
Routing and Providers
The Go backend is the gateway, scheduler, ledger, and task center. A generation request is authenticated, queued, routed to the active upstream provider configuration, billed, recorded in usage, and optionally sent to your webhook. Runtime provider routes are managed through the admin system. At the time this documentation was rewritten, the published public catalog contains only nano-banana and the runtime route table is empty, so docs should not advertise other public models.Parameters
Use
nanobanana.Prompt text. Minimum 1 character, maximum 2000 characters.
Optional text describing what to avoid. Maximum 1000 characters.
16:9 or 9:16. Defaults to 16:9.Number of output images, from
1 to 4. Defaults to 1. Billing multiplies by the number of returned images.Optional
1k, 2k, or 4k. The backend accepts this field; public nano-banana usage should treat it as provider-dependent.Supplying one or more image URLs switches the request into
image_edit mode. Up to 9 URLs are accepted.Optional HTTPS callback URL. EggAPI sends
task.completed or task.failed events after the worker finishes.Model Guide
nanobanana
Image generation and image editing with the public nano-banana model.