so, there is an issue where everything freezes out To be able to. Reproduce this bug. If you have a list of a conversation on the left panel, send any message to the agent and fast in a very fast way go to switch to another conversation. The Antigravity agent will stop. freezing, however it works in the background for some reason, I don’t know I tried to refresh reload Try to open the console. There is nothing that shows anything. I have to close it, totally and then reopen it again. To be able to continue the work. This happens a lot and it just our operation and specially with users like us, who is using it in a fast way
Hi @Yasin_Hassanien,
We haven’t been able to reproduce this unexpected behavior based on your description. To help us better assist you and investigate this further, could you please provide a few additional details?
- The specific version of Antigravity 2.0 you are using
- Your operating system
- A sample prompt or the type of task you are instructing the agent to perform
- Whether this is happening across all conversations or only specific ones (e.g., long conversations with extensive context)
- Whether you are experiencing this behavior across all models or only with specific ones
So i believe this happens becoase the recent convo jumps up in the order because it becomes a recent convo and when i hit enter then move to another covo then it freezes so i think you guys should delay the order movement till the agent finish the turn then moves up freely to avoid thsi
Window freezes from stuck renderer process (~100% CPU)
Issue: Every so often, a single Antigravity window freezes completely. The rest of the app and all other windows keep working fine. When it happens, one renderer process is pinning a CPU core at ~100–160% in a busy loop.
Key finding: The freeze is isolated to one Helper (Renderer) child process. The main Antigravity process stays at normal idle CPU — it’s never the culprit.
Recovery without restarting: Killing just that one stuck renderer fixes it. Antigravity automatically reloads the affected window, and everything else is untouched. No need to quit the app and lose every window. (Process tree: killing the stuck child → main app stays alive → fresh renderer spawns → window reloads.)
Possible trigger: The logs show the language server repeatedly failing to exit gracefully (getting SIGKILLed) plus frequent auto-restart / port-changed events, which seem to correlate with the freezes.
Suggestions for the team:
- A renderer CPU watchdog that auto-reloads a window when its renderer pins a core for too long.
- Capture a stack/minidump from a hung renderer before reloading, to find the loop.
- Harden the language-server shutdown/reload path (looks like a likely trigger).
Anyone else seeing single-window freezes? Would love a built-in “reload this window” option like a browser tab, so we don’t have to restart the whole app.
@Siddharth_Naik this became a big issue for me
I have been struggling with this issue for a while.
Although with mine, its not just the ui thread, but the main antigravity thread as it is too busy to even start up the automatic recovery.
To fix mine, I had to both close the ui thread in task manager, and hard reload the window graphics using ctrl + win + r.
At least the work around lets you continue without losing work or active tasks, but it is very annoying. I’ve noticed that it happens fairly frequently if you have long running tasks with large logging outputs.