Skip to main content
GET
/
v1
/
models
/
{model_id}
Retrieve Model
curl --request GET \
  --url https://api.example.com/v1/models/{model_id}
{
  "id": "<string>",
  "provider": "<string>",
  "created_at": "<string>",
  "display_name": "<string>",
  "description": "<string>",
  "version": "<string>",
  "capabilities": {
    "text": true,
    "vision": true,
    "audio": true,
    "image_generation": true,
    "tools": true,
    "structured_output": true,
    "streaming": true,
    "thinking": true,
    "input_token_limit": 123,
    "output_token_limit": 123
  },
  "defaults": {
    "temperature": 123,
    "max_output_tokens": 123,
    "top_p": 123,
    "top_k": 123
  },
  "provider_declared_generation_methods": [
    "<string>"
  ],
  "provider_info": {}
}

​
Method

client.models.retrieve(modelId)
Retrieve detailed information about a specific model, including its capabilities, provider, and supported features.

​
Parameters

​
modelId
string
required
The ID of the model to retrieve (e.g., "openai/gpt-4", "anthropic/claude-3-5-sonnet-20241022").

​
Response

​
id
string
required
Unique model identifier with provider prefix (e.g., "openai/gpt-4").
​
provider
string
required
Provider that hosts this model. One of: openai, anthropic, google, xai, mistral, groq, fireworks, deepseek.
​
created_at
string
required
When the model was released (RFC 3339 format).
​
display_name
string
Human-readable model name.
​
description
string
Model description.
​
version
string
Model version identifier.
​
capabilities
object
Normalized model capabilities across all providers.
​
defaults
object
Provider-declared default parameters for model generation.
​
provider_declared_generation_methods
string[]
Provider-specific generation method names.
​
provider_info
object
Raw provider-specific metadata.

​
Example

​
Basic retrieval

const model = await client.models.retrieve("openai/gpt-4");

console.log(`Model: ${model.id}`);
console.log(`Provider: ${model.provider}`);
console.log(`Display Name: ${model.display_name}`);

​
Check model capabilities

const model = await client.models.retrieve("anthropic/claude-3-5-sonnet-20241022");

if (model.capabilities?.vision) {
  console.log("This model supports vision!");
}

if (model.capabilities?.tools) {
  console.log("This model supports function calling");
}

console.log(`Max input tokens: ${model.capabilities?.input_token_limit}`);
console.log(`Max output tokens: ${model.capabilities?.output_token_limit}`);

​
Get model defaults

const model = await client.models.retrieve("openai/gpt-4");

if (model.defaults) {
  console.log(`Default temperature: ${model.defaults.temperature}`);
  console.log(`Default max tokens: ${model.defaults.max_output_tokens}`);
}

​
Response Example

{
  "id": "openai/gpt-4",
  "provider": "openai",
  "created_at": "2023-06-27T00:00:00Z",
  "display_name": "GPT-4",
  "description": "Most capable GPT-4 model for complex tasks",
  "version": "gpt-4-0613",
  "capabilities": {
    "text": true,
    "vision": false,
    "audio": false,
    "image_generation": false,
    "tools": true,
    "structured_output": true,
    "streaming": true,
    "thinking": false,
    "input_token_limit": 8192,
    "output_token_limit": 4096
  },
  "defaults": {
    "temperature": 1.0,
    "max_output_tokens": 4096,
    "top_p": 1.0
  },
  "provider_declared_generation_methods": [
    "generateContent",
    "streamGenerateContent"
  ],
  "provider_info": {
    "openai_specific_field": "value"
  }
}

​
Error Responses

​
404 Not Found

Returned when the model ID does not exist or is not accessible with your current API key. 
{
  "error": {
    "message": "Model not found",
    "type": "invalid_request_error",
    "code": "model_not_found"
  }
}

​
401 Unauthorized

Returned when authentication fails. 
{
  "error": {
    "message": "Invalid API key",
    "type": "invalid_request_error",
    "code": "invalid_api_key"
  }
}

Build docs developers (and LLMs) love

Get started for freeTalk to us