Skip to main content
Qwen Cloud home page
Docs
Clients & tools

Cursor

AI-powered code editor

Cursor is an AI-powered code editor that accelerates development with intelligent code completion and chat. Connect it to Qwen Cloud's pay-as-you-go API for direct access to powerful Qwen models through the OpenAI-compatible endpoint.

Quick start

Get running in a few minutes: 
# 1. Install
Download Cursor from cursor.com (Pro subscription required)

# 2. Configure (open Cursor Settings > Models tab)
OpenAI API Key: sk-xxx
Override OpenAI Base URL: https://dashscope-intl.aliyuncs.com/compatible-mode/v1

# 3. Test (add and use a model)
In "Add or search model" field, type: qwen3.5-plus
Click "Add Custom Model"
Select the model in chat panel and ask: "Explain what a REST API is"
You should see: The model appears in dropdown and provides a detailed explanation of REST API

Configuration

Basic setup

Configure Cursor to use Qwen Cloud:
  • API endpoint: https://dashscope-intl.aliyuncs.com/compatible-mode/v1
  • Authentication: API key required
  • Model selection: Add any Qwen model name manually
Cursor Pro or higher is required. The free plan supports only Auto mode and does not allow custom models.
For Coding Plan users: Use your exclusive base URL and API key instead. See Coding Plan for Cursor.

Step-by-step configuration

1

Open Cursor Settings

Click the settings icon icon → Cursor Settings → Models
2

Enter your API key

Enable OpenAI API Key and enter your API key
3

Set the base URL

Enable Override OpenAI Base URL and enter:
https://dashscope-intl.aliyuncs.com/compatible-mode/v1
4

Add models

In Add or search model, type the model name (e.g., qwen3.5-plus) → Add Custom Model
5

Start coding

Select your model in the chat panel and begin
Chat panel model selection

Model recommendations

TaskRecommended modelsWhy
Complex algorithmsqwen3.5-plus, qwen3-max, qwen3-coder-plusStrong reasoning and architecture design
Daily codingqwen3-coder-next, qwen3.5-flashFast responses, good for completion
Code reviewqwen3.5-plus, qwen3-coder-plusBetter understanding of context

Limitations

  • Requires: Cursor Pro subscription or higher
  • Does not support: Vision models (Qwen-VL) due to Cursor limitations
  • Note: Disable Auto mode to see custom models in dropdown

Examples

Code completion

# Type a comment and let Cursor complete the implementation
# Calculate fibonacci sequence
def fibonacci(n):
  # Cursor will auto-complete with Qwen model
  pass

Chat assistance

Chat: "Refactor this function to use async/await pattern"
Model: qwen3-coder-next
[Paste your code]

Architecture guidance

Chat: "Design a REST API for user authentication with JWT"
Model: qwen3.5-plus

Troubleshooting

"Cannot use added models" error
Solution: Upgrade to Cursor Pro. Free plans only support Auto mode, not custom models.
Model doesn't appear in dropdown
Solution: Disable Auto mode in the chat panel. Custom models only show when Auto is off.
"Trouble connecting to model provider"
Solution:
  • Verify the model name exists in Model list
  • Check API key is valid and has quota
  • Ensure base URL is exactly: https://dashscope-intl.aliyuncs.com/compatible-mode/v1
Slow responses
Solution:
  • Use faster models: qwen3-coder-next or qwen3.5-flash
  • Start new chat to reset context
  • Check network connection stability
Conflict with other providers
Solution: After configuring Qwen Cloud, requests to OpenAI fail. To switch back, disable both custom settings.