void icon indicating copy to clipboard operation
void copied to clipboard

How to enable autoComplete(TAB) ? I also have some other questions.

Open rfrkk opened this issue 8 months ago • 1 comments

version : 1.99.30034 (system setup) : 1.3.10 date : 906502f6600146442d41bc314c2dd011c686941f Electron: 2025-05-16T20:35:23.913Z ElectronBuildId: 34.3.2 Chromium: undefined Node.js: 132.0.6834.210 V8: 20.18.3 OS: 13.2.152.41-electron.0


Hello,

I am a Cursor user, trying to switch to Void AI. I've encountered some issues over the past few days and would appreciate your assistance.

System Environment

My system environment is: Windows 10 IoT LTSC 21H2, CPU 13500H, 32GB RAM, 2TB SSD.

Installed Model

openrouter paid gemini paid openai paid

Chat AI Panel Issues

  • I have successfully set Ctrl+I to open the Chat AI panel. How can I disable or change its keyboard shortcut?
  • Is there a way to prevent the Chat AI panel from appearing automatically when Void AI starts?

Severe Freezing Issues with Void AI

  • After every launch, Void AI freezes for approximately one minute, making it unresponsive to mouse and keyboard input (CPU usage is 1-3% during this time).
  • Similarly, after importing a workspace, Void AI freezes for about one minute, also becoming unresponsive (CPU usage is 1-3%).
  • When working in a maximized window, I frequently experience freezing. The only way to resolve this is by minimizing and then re-maximizing the window.

Maximized Window Display Issue

  • While using Void AI in a maximized window, the display often becomes corrupted or shows artifacts. The only way to fix this is by toggling the window out of and back into maximized mode.

TAB Auto-completion Feature

  • The TAB key does not trigger any auto-completion functionality. There is no response.What model do I need to install?

Thank you for your help.

rfrkk avatar May 29 '25 22:05 rfrkk

I use ollama for auto-completion. Regarding your question, qwen2.5-coder is suitable for auto-completion (https://ollama.com/library/qwen2.5-coder ). But you can also add "supportsFIM": true in the settings of any model, and it might work.

hardWorker254 avatar Jun 01 '25 07:06 hardWorker254

I also encountered qwen2.5-coder-ollama and API that did not respond for tab

mj520 avatar Jul 04 '25 09:07 mj520