Hey, this looks like a temporary issue on the provider side. resource_exhausted usually means the model provider was overloaded at the time of your request.
Can you try again now? It might have resolved itself.
If it’s still happening:
- Cursor Settings > Tools & MCP: if you have any MCP servers connected, try disabling them temporarily
- Cursor Settings > Models: make sure “Override OpenAI Base URL” is turned off, and disable any third-party API keys if you have them
- Sign out and sign back in (Cursor Settings > Sign Out)
This has happened before and was fixed using the steps above:
- Connection problems with various models despite clean network diagnostics and having tried to switch to HTTP/1.1
- Persistent Connection Error
Let me know how it goes.