Skip to main content

Models

Jupyter AI Agents can be used with multiple AI models providers.

UI​

Read the details on the Agent Runtimes Models documentation to learn how to configure and use different models in the Jupyter AI Agents UI.

CLI​

Azure OpenAI​

Jupyter AI Agents supports models from Azure OpenAI.

Read the Azure Documentation to get the needed credentials and make sure you define them in the following environment variables.

export OPENAI_API_VERSION="..."
export AZURE_OPENAI_ENDPOINT="..."
export AZURE_OPENAI_API_KEY="..."

Anthropic​

Jupyter AI Agents supports Claude models from Anthropic.

To use Anthropic Claude directly, you'll need to:

  1. Get an API key from Anthropic's Console
  2. Set up your environment variables:
export ANTHROPIC_API_KEY="..."

Example usage with direct Anthropic integration:

# Prompt agent example with Anthropic Claude
jupyter-ai-agents prompt \
  --url http://localhost:8888 \
  --token MY_TOKEN \
  --model-provider anthropic \
  --model-name claude-3-5-sonnet-20241022 \
  --path notebook.ipynb \
  --input "Create a matplotlib example"

You can use any of the following Claude models:

  • claude-3-haiku-20240307 - Fastest and most compact Claude model
  • claude-3-sonnet-20240229 - Balanced model for most tasks
  • claude-3-opus-20240229 - Most powerful Claude model
  • claude-3-5-sonnet-20241022 - Previous most intelligent Claude 3.5 model
  • claude-3-7-sonnet-20250219 - Most intelligent model Claude 3.7 model (recommended)

OpenAI​

Jupyter AI Agents supports models from OpenAI.

To use OpenAI directly, you'll need to:

  1. Get an API key from OpenAI
  2. Set up your environment variables:
export OPENAI_API_KEY="..."

Example usage with direct OpenAI integration:

# Prompt agent example with OpenAI
jupyter-ai-agents prompt \
  --url http://localhost:8888 \
  --token MY_TOKEN \
  --model-provider openai \
  --model-name gpt-4o-mini \
  --path notebook.ipynb \
  --input "Create a matplotlib example"

GitHub Copilot​

Jupyter AI Agents supports models from GitHub Copilot.

Read the Langchain GitHubCopilot README to get the GITHUB token and make sur you define it in the following environment variable.

export GITHUB_TOKEN="..."

GitHub Copilot supports different models that you can specify using the --model-name parameter. The available models as of 2024-02-07 are

  • gpt-4o
  • o1
  • o3-mini

Make sure you have a GitHubCopilot subscription and have enabled the model you want to use in your GitHubCopilot settings.