API Models

Use GET /api/v1/models to list Rivya models with their Public API readiness.

curl https://rivya.ai/api/v1/models

The endpoint is public and read-only. It returns the same public model layer used by the API docs and online debugger, including models that are not fully callable yet.

Response Shape

The response uses this shape:

{
  "object": "list",
  "data": [
    {
      "id": "z-image",
      "display_name": "Z-Image",
      "category": "image",
      "runtime": "async_task",
      "base_credits": 1,
      "billing_type": "FIXED",
      "supported_modes": ["text-to-image"],
      "max_prompt_length": 1000,
      "reference_media": null,
      "params": [],
      "api_status": "available",
      "api_input_status": "available_text",
      "supported_api_inputs": ["text"],
      "requires_files_api": false,
      "requires_upload_for_modes": [],
      "url_input_supported": false,
      "file_kinds": [],
      "chat_capabilities": null,
      "notices": []
    }
  ]
}

Use id as the model value in POST /api/v1/generations for asynchronous generation models. For chat models, use POST /api/v1/chat/completions and check chat_capabilities.

Availability States

The model reference uses these public states:

• available: at least one Public API generation path is available

• temporarily_unavailable: model is temporarily unavailable and cannot currently be called through Public API v1. Read-only entries do not expose runtime, billing, pricing, modes, parameters, or chat capabilities; choose an available model instead.

• studio_only: model is available through Studio or another product area, not through Public API v1

The input readiness field is more specific:

• available_text: prompt-based requests can be tested directly

• available_url_input: public URL or scalar params can be tested directly

• partial_requires_files: some modes can be called without upload; reference-media modes should use Files API

• requires_files_api: the model depends on uploaded reference media through /api/v1/files

Use Files API before calling modes that require uploaded image, video, or audio references.

For chat models, chat_capabilities documents whether Chat API supports non-streaming responses, streaming responses, web search, image file_id attachments, thoughts, and reasoning effort.

Model Count Policy

The public model list changes as Rivya adds, removes, or gates models.

Do not rely on a hard-coded model count in your integration. Always read /api/v1/models or the dynamic Model API Reference.

Related Pages

• Model API Reference

• Create Generation

• Chat API

• Rivya Model Selection Guide

• Model Availability in Rivya

Coming Soon discovery entries

• coming_soon: the model is listed for discovery but cannot be submitted to generation or Chat API. These read-only entries return supports_direct_generation: false, null runtime, billing, and pricing fields, plus empty supported_modes and params. Choose a model with api_status: available for requests.