---
title: "Fetch Models"
method: GET
path: "/v1/models"
tags: ["Models"]
---

# Fetch Models

`GET /v1/models`

## Request

- Base URL: `https://apipie.ai`
- URL: `https://apipie.ai/v1/models`
- Auth: API key in header `Authorization`

Fetch all available models or use one or more filters

## Query parameters

- `type` 'llm' | 'vision' | 'embedding' | 'image' | 'voice' | 'moderation' | 'coding' | 'free'
- `subtype` 'chat' | 'fill-mask' | 'question-answering' | 'tts' | 'stt' | 'multimodal'
- `provider` string
- `combination` string
- `enabled` 1 | 0
- `voices` unknown
- `restrictions` string

## Response `200`

Models or voices list depending on query parameter used.

- union
  - ModelResponse
    - `object` string: Type of the list
    - `data` Model[]
      - `enabled` boolean, required: Whether Model is enabled
      - `type` 'llm' | 'vision' | 'embedding' | 'image' | 'voice' | 'moderation' | 'code', required: Type of the model
      - `subtype` string: Subtype of the model
      - `provider` string, required: Provider of the model
      - `model` string, required: The base model name
      - `route` string, required: Provider specific identifier
      - `description` string: Description of the Model
      - `max_tokens` integer, required: Max cumulative tokens supported by the model
      - `max_response_tokens` integer, required: Max tokens that this model can return in the response
      - `req_price` string: Price for request - per unit. If not present, price only provided by the provider in query response
      - `res_price` string: Price for response - per unit. If not present, price only provided by the provider in query response
      - `price_unit` string: Price unit, the number of price_type per price
      - `price_type` 'token' | 'char' | 'request' | 'tokens' | 'image' | 'minute' | 'characters' | 'character' | 'provider': Type of the price
      - `avg_cost` string: Average cost of the model, tracking actual usage. This is the average cost per 1000 characters
      - `latency` string, required: Average latency by combined prompt/response size, in the format: 2000/4000/8000/16000/32000/Average_Latency_per_1000_characters
      - `available` boolean, required: Availability status of the model
      - `query_count` string, required: Total count of queries for this model
  - VoiceModelResponse
    - `object` string: Type of the response object, always a list.
    - `data` VoiceModel[]
      - `provider` string: Provider of the voice model.
      - `model` string: Model identifier.
      - `voice` string: Unique identifier for the voice.
      - `name` string: Name of the voice.
      - `description` string: Description of the voice characteristics.
  - RestrictionsResponse
    - `object` string: Type of the response object, always a list.
    - `data` Restrictions[]
      - `restrict_id` integer: Unique identifier for the restrict table.
      - `name` string: Name or description, usually the provider name.
      - `countries` string: JSON string of supported countries.

---

[API](https://skmtc.dev/apipie/apis/introduction.md) · [All operations](https://skmtc.dev/apipie/apis/introduction/llms.txt) · [OpenAPI document](https://skmtc.dev/apipie/apis/introduction/revisions/f8386be4eb3f?raw)