List Expert Models

Use the /v3/info endpoints to discover all available expert model features, providers, and models.

Endpoints

Endpoint	Description	Query Parameters
`GET /v3/info`	List all features and subfeatures	`include=schemas`, and with it `format=simplified` or `format=json_schema`
`GET /v3/info/{feature}`	List the subfeatures of one feature	None
`GET /v3/info/{feature}/{subfeature}`	Get providers, models, pricing, and schemas for a specific feature	`format=simplified` or `format=json_schema`
`GET /v3/info/audio/tts/voices`	List the voices the text-to-speech models accept	`model`, `language`, `gender`

List All Features

curl https://api.edenai.run/v3/info \
  -H "Authorization: Bearer YOUR_API_KEY"

import requests

response = requests.get(
    "https://api.edenai.run/v3/info",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

for feature in response.json()["features"]:
    print(feature["name"])

const response = await fetch("https://api.edenai.run/v3/info", {
  headers: { "Authorization": "Bearer YOUR_API_KEY" }
});

const { features } = await response.json();
features.forEach(f => console.log(f.name));

Bulk Schema Discovery

To avoid querying each subfeature individually, pass ?include=schemas to embed full schema bundles across all subfeatures in a single response:

curl "https://api.edenai.run/v3/info?include=schemas" \
  -H "Authorization: Bearer YOUR_API_KEY"

import requests

response = requests.get(
    "https://api.edenai.run/v3/info?include=schemas",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

for feature in response.json()["features"]:
    for subfeature in feature["subfeatures"]:
        fields = [field["name"] for field in subfeature["input_schema"]["fields"]]
        print(f'{feature["name"]}/{subfeature["name"]}:', ", ".join(fields))

With include=schemas, each subfeature also carries the fields GET /v3/info/{feature}/{subfeature} returns: endpoints, request_schema, input_schema, output_schema, and provider_params. Add format=json_schema to get full JSON Schemas instead of the default simplified ones. format only applies to these schemas, so without include=schemas it has no effect.

HTTP Caching and Conditional Requests

Every /v3/info route returns a weak ETag header alongside Cache-Control: private, no-cache. Keep the ETag of a response and send it back in an If-None-Match header: while the response is unchanged, the API answers 304 Not Modified with an empty body, so polling the catalog costs no download.

import requests

url = "https://api.edenai.run/v3/info"
headers = {"Authorization": "Bearer YOUR_API_KEY"}

first = requests.get(url, headers=headers)
etag = first.headers["ETag"]

# Later: ask again, sending back the ETag you kept
second = requests.get(url, headers={**headers, "If-None-Match": etag})
print(second.status_code)  # 304 while the catalog is unchanged

const url = "https://api.edenai.run/v3/info";
const headers = { "Authorization": "Bearer YOUR_API_KEY" };

const first = await fetch(url, { headers });
const etag = first.headers.get("ETag");

// Later: ask again, sending back the ETag you kept
const second = await fetch(url, { headers: { ...headers, "If-None-Match": etag } });
console.log(second.status); // 304 while the catalog is unchanged

Get Feature Details

Retrieve providers, pricing, endpoint paths, request/response schemas, and parameter allowlists for a specific feature.

curl https://api.edenai.run/v3/info/text/moderation \
  -H "Authorization: Bearer YOUR_API_KEY"

import requests

response = requests.get(
    "https://api.edenai.run/v3/info/text/moderation",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

details = response.json()
for model in details.get("models", []):
    print(model["model"], model["pricing"])

Response Fields

Subfeature Details

Field	Description
`feature`, `subfeature`	Names used in the `model` string (e.g. `text` and `moderation`)
`feature_fullname`, `subfeature_fullname`	Display names
`description`	What the subfeature does
`mode`	`sync` for subfeatures you call with `POST /v3/universal-ai`, `async` for those you call with `POST /v3/universal-ai/async` and then poll
`endpoints`	The operations you can call, each as `METHOD /path`: `create` for a sync subfeature, `create`, `get`, `list`, and `delete` for an async one. Subfeatures backed by collections (`image/face_recognition`) add `create_collection`, `add_item`, and the other collection operations
`models`	List of available models with pricing, documentation links, and input constraints
`request_schema`	Top-level request body schema expected by the endpoint, showing where inputs nest
`input_schema`	Schema of the parameters accepted inside the `input` object
`output_schema`	Schema of the standardized response returned inside the `output` object
`provider_params`	For each provider, the `provider_params` it accepts on subfeatures that restrict them, or `null` on the others. See Provider Parameters

Model Metadata

Each entry in the models list contains:

Field	Description
`model`	Full model identifier string (e.g. `text/moderation/google`)
`pricing`	Cost information including `price`, `price_unit_quantity`, `price_unit_type` (such as `request`, `second`, `execution_time`, `character`), and `currency` (`USD`)
`regions`	Regions the model is served from, as `code` and `name` objects (e.g. `eu` and `Europe`); empty when no region is recorded for the model
`documentation_url`	Direct URL to official provider documentation for the model or provider, or `null`
`requires_reference_images`	Boolean flag indicating whether the model requires `reference_images` (e.g. image edit models)
`requires_video`	Boolean flag indicating whether the model requires a source `video` input (e.g. video editing models)
`zero_data_retention`	`true` when the provider keeps no inputs or outputs after the response (Zero Data Retention), `false` when it keeps them, `null` when not verified yet

You can also browse all features and providers visually in the Eden AI model catalog.

Need models whose provider keeps no data? Add ?zero_data_retention=true to /v3/info, /v3/info/{feature} or /v3/info/{feature}/{subfeature} to list only the model entries whose zero_data_retention is true.

Need EU data residency? Hit https://api.eu.edenai.run/v3/info instead and the response is automatically filtered to EU-eligible providers and models. See EU Endpoint.

Looking for LLMs like GPT-4o or Claude? See List LLM Models for the full list of chat models.

V3 Documentation

Quick Start

Overview

LLMs

Expert Models

General

Organization

Data Governance

Integrations

Endpoints

List All Features

Bulk Schema Discovery

HTTP Caching and Conditional Requests

Get Feature Details

Response Fields

Subfeature Details

Model Metadata

​Endpoints

​List All Features

​Bulk Schema Discovery

​HTTP Caching and Conditional Requests

​Get Feature Details

​Response Fields

​Subfeature Details

​Model Metadata

Endpoints

List All Features

Bulk Schema Discovery

HTTP Caching and Conditional Requests

Get Feature Details

Response Fields

Subfeature Details

Model Metadata