Hey , guys!
I’m excited about iTerm2’s AI inline autocompletion feature—it worked almost out of the box with minimal configuration and is far superior to alternatives like Warp and Amazon Q. However, I’m experiencing some issues:
1. On my second Mac, inline autocompletion simply isn’t working.
2. After changing the AI model on my primary Mac, the feature stopped working altogether.
To troubleshoot, I have:
• Uninstalled and reinstalled both iTerm2 and the AI plugin.
• Experimented with different AI models and API endpoints (using services like OpenRouter and OpenAI).
• Adjusted settings such as legacy completion modes, as well as configurations for chat completions and chat responses.
• Removed advanced settings related to AI model name prefixes.
Notably, in some configurations the Shift+Command+Dot shortcut successfully sends a request and receives an appropriate API response. This indicates that the API connectivity is intact, and the issue appears isolated to the inline autocompletion itself.
Could you please provide any insight or suggestions on how to resolve this issue? I suspect it might be a minor configuration problem that I’m overlooking, but I haven’t been able to pinpoint it.
Thank you in advance for your assistance. Please let me know if you need any further information or logs.