Models API

The Models API allows you to list and retrieve information about the models available through Conduit.

List Models

Lists the models available for use through Conduit.

Endpoint

GET /v1/models

Headers

Header	Value	Required	Description
Authorization	Bearer condt_your_virtual_key	Yes	Your Conduit virtual key

Query Parameters

Parameter	Type	Required	Description
capability	string	No	Filter by model capability (e.g., 'chat', 'embeddings')
provider	string	No	Filter by provider (e.g., 'openai', 'anthropic')

Example

cURL

curl http://localhost:5000/v1/models \
  -H "Authorization: Bearer condt_your_virtual_key"

Python

from openai import OpenAI

client = OpenAI(
    api_key="condt_your_virtual_key",
    base_url="http://localhost:5000/v1"
)

models = client.models.list()
for model in models.data:
    print(model.id)

JavaScript

import OpenAI from 'openai';

const openai = new OpenAI({
  apiKey: 'condt_your_virtual_key',
  baseURL: 'http://localhost:5000/v1'
});

async function main() {
  const models = await openai.models.list();
  models.data.forEach(model => {
    console.log(model.id);
  });
}

main();

C#

using System;
using System.Net.Http;
using System.Threading.Tasks;

class Program
{
    static async Task Main()
    {
        var client = new HttpClient();
        client.BaseAddress = new Uri("http://localhost:5000/");
        client.DefaultRequestHeaders.Add("Authorization", "Bearer condt_your_virtual_key");

        var response = await client.GetAsync("v1/models");
        var responseBody = await response.Content.ReadAsStringAsync();
        
        Console.WriteLine(responseBody);
    }
}

Response Format

{
  "object": "list",
  "data": [
    {
      "id": "my-gpt4",
      "object": "model",
      "created": 1677610602,
      "owned_by": "conduit",
      "provider": "anthropic",
      "provider_model": "claude-3-opus-20240229",
      "capabilities": ["chat", "function_calling", "vision"]
    },
    {
      "id": "my-embedding-model",
      "object": "model",
      "created": 1677649963,
      "owned_by": "conduit",
      "provider": "openai",
      "provider_model": "text-embedding-ada-002",
      "capabilities": ["embeddings"]
    }
  ]
}

Retrieve Model

Retrieves a specific model's information.

Endpoint

GET /v1/models/{model_id}

Example

cURL

curl http://localhost:5000/v1/models/my-gpt4 \
  -H "Authorization: Bearer condt_your_virtual_key"

Python

from openai import OpenAI

client = OpenAI(
    api_key="condt_your_virtual_key",
    base_url="http://localhost:5000/v1"
)

model = client.models.retrieve("my-gpt4")
print(model)

JavaScript

import OpenAI from 'openai';

const openai = new OpenAI({
  apiKey: 'condt_your_virtual_key',
  baseURL: 'http://localhost:5000/v1'
});

async function main() {
  const model = await openai.models.retrieve("my-gpt4");
  console.log(model);
}

main();

Response Format

{
  "id": "my-gpt4",
  "object": "model",
  "created": 1677610602,
  "owned_by": "conduit",
  "provider": "anthropic",
  "provider_model": "claude-3-opus-20240229",
  "capabilities": ["chat", "function_calling", "vision"],
  "context_length": 200000,
  "status": "available"
}

Conduit Extensions

Conduit includes additional model-related endpoints:

List Provider Models

Lists all available models from a specific provider.

GET /v1/provider-models/{provider}

cURL

curl http://localhost:5000/v1/provider-models/openai \
  -H "Authorization: Bearer condt_your_virtual_key"

Model Capabilities

The capabilities field in model responses indicates what features a model supports:

Capability	Description
chat	Supports chat completions
completions	Supports text completions
embeddings	Supports text embeddings
function_calling	Supports function calling / tools
vision	Supports image inputs
streaming	Supports streaming responses
json_mode	Supports JSON mode output

Model Status

The status field can have these values:

Status	Description
available	Model is available for use
unavailable	Model is temporarily unavailable
deprecated	Model is deprecated and may be removed
limited	Model has usage limitations

Best Practices

Virtual Model Abstraction

Conduit virtual models abstract away specific provider models. This means:

• Your applications should use the virtual model names (my-gpt4)
• The underlying provider model can be changed without affecting your code
• Different provider models can be grouped under the same virtual model name

Model Discovery Workflow

1. List all available models with the Models API
2. Filter models by capability for your use case
3. Test models to find the best balance of performance vs. cost
4. Configure model mappings in Conduit for flexibility

Error Handling

HTTP Code	Error Type	Description
401	authentication_error	Invalid or missing API key
403	permission_error	The API key doesn't have permission
404	not_found_error	The requested model was not found
429	rate_limit_error	Rate limit exceeded
500	server_error	Server error

Next Steps

• Chat Completions API: Generate conversational responses
• Embeddings API: Generate vector embeddings
• Model Routing: Learn about advanced routing options