Skip to main content

​
GET /api/models

Retrieves the list of available Ollama models for script generation and the default model to use.

​
Response

​
status
string
"success" if models were retrieved successfully, "error" if Ollama is not running or unreachable.
​
models
array
Array of available model names.Example: ["llama3.1:8b", "llama3.1:70b", "mistral:7b"]
​
default
string
The default model name to use for generation.Example: "llama3.1:8b"
​
message
string
Only present on error responses. Describes why model fetching failed.

​
Example Request

curl http://localhost:8080/api/models

​
Example Response (Success)

{
  "status": "success",
  "models": [
    "llama3.1:8b",
    "llama3.1:70b",
    "mistral:7b",
    "phi3:mini"
  ],
  "default": "llama3.1:8b"
}

​
Example Response (Error)

When Ollama is not running or unreachable: 
{
  "status": "error",
  "message": "Could not fetch Ollama models. Is Ollama running?",
  "models": [
    "llama3.1:8b"
  ],
  "default": "llama3.1:8b"
}
Even on error, the response includes a fallback model list from the OLLAMA_MODEL environment variable.

​
Usage

The models returned by this endpoint can be used in the aiModel field when creating a generation job: 
import requests

# Get available models
models_response = requests.get("http://localhost:8080/api/models")
default_model = models_response.json()["default"]

# Use in generation request
requests.post(
    "http://localhost:8080/api/generate",
    json={
        "videoSubject": "AI technology",
        "aiModel": default_model,  # Use the default model
        "paragraphNumber": 3
    }
)

​
Model Selection

Performance vs Quality Trade-off
  • Smaller models (e.g., llama3.1:8b, phi3:mini) are faster but may produce simpler scripts
  • Larger models (e.g., llama3.1:70b) produce higher quality scripts but take longer
  • The default model is usually a good balance for most use cases

​
Troubleshooting

If the endpoint returns an error:
  1. Check if Ollama is running: 
    ollama list
  2. Verify the Ollama server is accessible: 
    curl http://localhost:11434/api/tags
  3. Check the OLLAMA_MODEL environment variable in .env: 
    OLLAMA_MODEL=llama3.1:8b
See the Ollama Models Guide for more details.

Build docs developers (and LLMs) love

Get started for freeTalk to us