↓
Rivya AI Docs

Rivya API Changelog

Read Rivya API v1 documentation notes, current endpoints, webhook limits, SDK beta status, and compatibility guidance.

Last reviewed on November 11, 2026

Use this page to check the API v1 documentation record, current endpoints, and compatibility notes.

Prompt units and complete layer results

Read prompt_length_unit together with max_prompt_length from GET /api/v1/models. utf8_bytes counts the original text bytes, including spaces and line breaks; utf16_code_units counts JavaScript UTF-16 units. A null limit does not mean unlimited or optional text. Gemini TTS uses the byte unit.

A completed layer-decomposition result can include result.layers alongside the complete result.urls list. Keep each layer’s z_index, URL, size, name, description and bounding box when provided; index zero is the base image. The TypeScript SDK exposes these fields. Upload references for the selected model and pass its original signed metadata; declared dimensions or duration alone do not replace verification.

Suno Speech: scenes and supplied scripts

Suno Speech: prompt is the scene in Simple mode (params.customMode="false") or the script in Advanced mode ("true"). Advanced-only settings are tone, variety, vocalGender and backgroundMusic. Omit them in Simple mode; do not send a second script or prompt inside params.

2026-05-11

This documentation entry is dated May 11, 2026. The list below describes the API v1 docs; it does not give launch dates for individual features.

  • API overview

  • quickstart

  • authentication

  • OpenAPI and schema contract

  • API model list

  • dynamic model API reference

  • generation submission

  • generation status polling

  • credit balance

  • errors and limits

  • Files API for reference uploads

  • signed webhook events for API-created generation tasks

  • Chat API responses, SSE streams, and API-created sessions

Current API v1 scope

API v1 covers asynchronous image, video, and audio generation, model reference uploads through Files API, and Chat responses returned as JSON or SSE. Creating a generation or Chat turn requires API access to be enabled for the deployment, a valid key with the required scopes, and a model available for that endpoint. Generation and Chat use the key owner's account credits.

Signed generation webhooks cover API-created tasks. generation.succeeded requires a successful task whose settlement_status is SETTLED; generation.failed applies to a failed task.

The webhook feature must be enabled, and the task's account must have an active endpoint. Delivery depends on that endpoint's event subscriptions. See API Webhooks for signatures and delivery rules. Studio tasks and Chat turns do not emit these generation events.

Current Endpoint Surface

The following routes are documented in API v1. Model discovery and OpenAPI are public and read-only; the other routes require an API key with the relevant scopes. Webhook management also requires the webhook feature to be enabled.

  • GET /api/v1/models

  • POST /api/v1/generations

  • POST /api/v1/files

  • GET /api/v1/files/{fileId}

  • POST /api/v1/chat/completions

  • POST /api/v1/chat/completions/stream

  • GET /api/v1/chat/sessions

  • GET /api/v1/chat/sessions/{sessionId}

  • GET /api/v1/generations/{taskId}

  • GET /api/v1/credits

  • GET /api/v1/webhooks

  • POST /api/v1/webhooks

  • GET /api/v1/webhooks/{endpointId}

  • PATCH /api/v1/webhooks/{endpointId}

  • DELETE /api/v1/webhooks/{endpointId}

  • POST /api/v1/webhooks/{endpointId}/rotate-secret

  • GET /api/v1/webhook-events

  • GET /api/v1/webhooks/{endpointId}/deliveries

  • POST /api/v1/webhooks/{endpointId}/test

  • GET /api/v1/openapi.json

Not Open Yet

The following limits still apply:

  • Chat webhook events are not supported.

  • The TypeScript SDK is a private beta and is not published as a public npm package. If you have access to the beta source, see the SDK guide for server-side integration. Otherwise, use the HTTP endpoints directly.

Compatibility Notes

The model list and availability can change. Read /api/v1/models or the Model API Reference for current IDs and availability instead of relying on a fixed model count.

New generation, Chat, and file-upload requests use different model lists in the OpenAPI schema. Historical generation and Chat responses keep a string model field so they can still return older IDs. Do not validate historical responses against the model enum for a new request.