Skip to main content
The AI Gateway exposes a /models endpoint that returns all models your API key has access to. This is useful for building dynamic model selectors, validating model IDs before making requests, and integrating with tools that auto-discover available models.

​
List All Models

Send a GET request to the /models endpoint: 
from openai import OpenAI

client = OpenAI(
    api_key="your_truefoundry_api_key",
    base_url="https://gateway.truefoundry.ai"
)

models = client.models.list()

for model in models.data:
    print(model.id)

​
Response Format

The endpoint returns an OpenAI-compatible response: 
{
  "object": "list",
  "data": [
    {
      "id": "openai-main/gpt-4o-mini",
      "object": "model",
      "owned_by": "openai-main"
    },
    {
      "id": "anthropic-main/claude-4-sonnet",
      "object": "model",
      "owned_by": "anthropic-main"
    },
    {
      "id": "tfy-ai-bedrock/us-anthropic-claude-sonnet-4-20250514-v1-0",
      "object": "model",
      "owned_by": "tfy-ai-bedrock"
    }
  ]
}
The response only includes models that your API key (PAT or VAT) has been granted access to. If a model is missing, check that the relevant provider account has been shared with your user or virtual account. See Access Control for details.

​
Use Cases

  • Dynamic model selectors: Build UIs that automatically populate model dropdowns from the API instead of hardcoding model IDs.
  • Validation: Verify that a model ID exists and is accessible before sending a request.
  • Tool integrations: Applications like n8n use the /models endpoint to auto-discover available models.
  • Inventory auditing: Programmatically audit which models are available across your organization.