Skip to main content
Cursor is an AI-powered code editor built on VSCode. By integrating Passy with Cursor, you gain access to multiple AI models, better cost control, and enhanced capabilities for your development workflow.

​
Prerequisites

Before you begin, ensure you have:

​
Integration Guide

Cursor supports OpenAI-compatible API endpoints, making it easy to integrate with Passy.
1

​
Get Your Passy API Key

First, you’ll need to obtain your Passy API key.
  1. Log in to your Passy dashboard
  2. Navigate to the API Keys section
  3. Create a new API key and copy it for later use
2

​
Configure Cursor Settings

Now let’s configure Cursor to use Passy as your AI provider.Cursor Settings Menu
  1. Open Cursor and press Cmd/Ctrl + , to open Settings
  2. Search for “Cursor Settings” or click on the Cursor tab
  3. Click on Models in the sidebar
3

​
Add Your API Key

Add OpenAI API Key
  1. In the Models section, scroll down to the OpenAI API Key section
  2. Click on Add OpenAI API Key
  3. Enter your Passy API key in the input field
  4. Click Save
4

​
Configure Base URL

Override Base URL
  1. In the same Models settings page, find the Override OpenAI Base URL option
  2. Enable the toggle for Override OpenAI Base URL
  3. Enter Passy’s API endpoint: https://api.passy.ai/v1
  4. Click Save
5

​
Select Your Models

Now you can select which models to use for different features.
  1. In the Models section, you’ll see options for:
    • Chat Model - for AI conversations (Cmd/Ctrl + L)
    • Autocomplete Model - for code suggestions as you type
  2. Choose any Passy supported model:
Popular model options:
  • gpt-5 - OpenAI’s GPT-5 (balanced performance)
  • claude-4.5-sonnet - Anthropic’s Claude 3.5 Sonnet (excellent for coding)
  • gemini-3.0-flash - Google’s Gemini 3.0 Flash (fast responses)
  • gpt-5-mini - Cost-effective for autocomplete
Advanced model formats:
Tip: Use a powerful model like claude-4.5-sonnet for chat, and a faster, cheaper model like gpt-5-mini for autocomplete to optimize cost and performance.
6

​
Test the Integration

Verify that everything is working correctly.
  1. Open any code file in Cursor
  2. Try the AI chat by pressing Cmd/Ctrl + L
  3. Ask a question like “Explain this code”
  4. Test autocomplete by starting to write code
Success! All AI requests are now routed through Passy. You can monitor usage and costs in your Passy dashboard.

​
Features

Once configured, you can use all of Cursor’s AI features powered by Passy:

​
AI Chat (Cmd/Ctrl + L)

Open an AI-powered chat panel to:
  • Ask questions about your codebase
  • Request code explanations and documentation
  • Get debugging help and suggestions
  • Generate new code snippets

​
Inline Edit (Cmd/Ctrl + K)

Edit code directly with natural language:
  • Refactor functions and classes
  • Add new features to existing code
  • Fix bugs with AI suggestions
  • Optimize and improve code quality

​
Autocomplete

Get intelligent code suggestions as you type:
  • Context-aware completions based on your codebase
  • Multi-line code predictions
  • Smart imports and function signatures
  • Adaptive learning from your coding patterns

​
Advanced Configuration

​
Using Different Models for Different Features

Optimize your workflow by using different models for different tasks:
FeatureRecommended ModelWhy
Chatclaude-3.5-sonnet or gpt-4oBest for complex reasoning and explanations
Autocompletegpt-4o-miniFast and cost-effective for real-time suggestions
Code Reviewgpt-4oThorough analysis and suggestions

​
Model Routing

Passy’s intelligent routing automatically:
  • Chooses cost-effective models by default for optimal price-to-performance
  • Scales to more powerful models based on your request’s complexity
  • Handles large contexts intelligently by selecting models with appropriate context windows
  • Falls back to alternatives if a model is unavailable

​
Troubleshooting

​
Authentication Errors

If you encounter authentication issues:
  • Verify your API key is correct and active
  • Ensure the base URL is set to https://api.passy.ai/v1
  • Check your Passy account has sufficient credits in the dashboard
  • Try removing and re-adding the API key in Cursor settings

​
Model Not Found

If you see “model not found” errors:
  • Verify the model ID exists in the models catalog
  • Check you’re using the correct model name format
  • Some provider-specific models may require additional configuration
  • Try using a standard model name like gpt-4o or claude-3.5-sonnet

​
Slow Responses

If responses are slower than expected:
  • Check your internet connection speed
  • Monitor your usage in the Passy dashboard
  • Consider using faster models like gpt-4o-mini for autocomplete
  • Reduce context size by limiting the number of files Cursor analyzes

​
Benefits of Using Passy with Cursor

​
Multi-Provider Access

Access models from OpenAI, Anthropic, Google, Meta, and more through a single API.

​
Cost Control

Track and limit your AI spending with detailed usage analytics and budget controls.

​
Response Caching

Reduce costs by up to 90% with intelligent caching of similar requests.

​
Analytics Dashboard

Monitor usage patterns, costs per model, and optimize your AI spend.

​
Automatic Fallbacks

If one provider is down, Passy automatically routes to alternatives.

​
Smart Routing

Let Passy choose the best model for each task based on cost and performance.

​
Next Steps

Now that you have Cursor integrated with Passy:

​
Need Help?

WhatsApp Support

Chat with us instantly for quick questions

Book a Call

Schedule a call for enterprise routing, custom GPU servers, or migration assistance