Skip to main content
Connect JetBrains AI Assistant to the Ollama API Proxy to use OpenAI, Google Gemini, and OpenRouter models through the Ollama API interface.

​
Prerequisites

  • JetBrains IDE (IntelliJ IDEA, PyCharm, WebStorm, etc.) with AI Assistant plugin installed
  • Ollama API Proxy running locally or on a remote server
  • At least one provider API key configured (OPENAI_API_KEY, GEMINI_API_KEY, or OPENROUTER_API_KEY)

​
Configuration

1

Start the Ollama API Proxy

Ensure the proxy server is running. By default, it runs on http://localhost:11434.
npm start
You should see output showing available providers and models:
🚀 Ollama Proxy with Streaming running on http://localhost:11434
🔑 Providers: openai, google, openrouter
📋 Available models: gpt-4o-mini, gpt-4.1-mini, gemini-2.5-flash, deepseek-r1
2

Open JetBrains AI Assistant settings

In your JetBrains IDE:
  1. Navigate to Settings (or Preferences on macOS)
  2. Go to Tools → AI Assistant
  3. Click on Model Provider
3

Configure Ollama as the provider

In the Model Provider settings:
  1. Select Ollama from the provider dropdown
  2. Set the Server URL to your proxy location:
    • Local: http://localhost:11434
    • Remote: http://your-server:11434
  3. Click Test Connection to verify the connection
If the proxy is running on a different port, update the URL accordingly. The port is configurable via the PORT environment variable.
4

Select a model

Once connected, you’ll see all available models from the proxy:
  • OpenAI models: gpt-4o-mini, gpt-4.1-mini, gpt-4.1-nano, gpt-4o, gpt-5-nano
  • Google models: gemini-2.5-flash, gemini-2.5-flash-lite
  • OpenRouter models: deepseek-r1, kimi-k2
Choose the model you want to use as the default for AI Assistant.
The available models depend on which API keys you’ve configured in your .env file. Models will only appear if their corresponding provider is available.
5

Start using AI Assistant

You’re all set! Use JetBrains AI Assistant features:
  • Code completion and suggestions
  • Chat with AI about your code
  • Explain code functionality
  • Generate tests and documentation
  • Refactor code with AI assistance
All requests will route through the Ollama API Proxy to your configured providers.

​
Switching between models

You can easily switch between different models:
  1. Open the AI Assistant chat panel
  2. Click the model selector at the top
  3. Choose a different model from the available options
This allows you to compare responses from different providers (OpenAI, Google, OpenRouter) without changing your configuration.

​
Troubleshooting

If the test connection fails:
  • Verify the proxy server is running (curl http://localhost:11434 should return “Ollama is running in proxy mode.”)
  • Check the server URL is correct
  • Ensure no firewall is blocking the connection
  • Check the proxy logs for any error messages
If no models appear in the selection:
  • Verify at least one provider API key is configured in your .env file
  • Check the proxy startup logs to see which providers are loaded
  • Ensure models.json exists and is properly formatted
  • Restart the proxy server after adding new API keys
If a selected model doesn’t respond:
  • Check the proxy server logs for API errors
  • Verify your API key for that provider is valid and has sufficient quota
  • Try a different model to isolate the issue
  • Check your network connection

​
Next steps

Supported Models

View all available models and their providers

Vision Support

Learn how to use vision models with images

Build docs developers (and LLMs) love

Get started for freeTalk to us