Cloud Agent (GitHub, Slack, Web, Linear)
Cannot use grok 4.7 model
use cursor cloud agent, second chat (after few hours) got error
first chat can use grok
should normal
Other
Cursor cloud agent
Yes - Cursor is unusable
me too
+1 affecting me too
+1 over here
Cloud Agent (GitHub, Slack, Web, Linear)
Run Test on a scheduled automation fails immediately with Failed to start background composer: [not_found] Error when the model is Grok 4.7. The same automation, same environment, starts when the model is Grok 4.6. The only change between the failing and working runs is the model.
Steps to Reproduce
Open an existing automation on cursor.com/automations. Trigger: custom schedule. Environment: environment is unchanged
Set the model to Grok 4.7.
Click Run Test.
The dialog shows Failed to start background composer: [not_found] Error.
Set the model to Grok 4.6 and click Run Test again. The run starts.
Run Test starts the background composer on Grok 4.7 the same way it does on Grok 4.6.
Other
Cursor Cloud, Web, no local clients involved.
Grok 4.6 (working) and Grok 4.7 (not working)
Settings: non fast and xhigh for both, for 4.7 we choose 500k context size.
None. The run never starts, so there is no bc-id. But see screenshot.
Automation in question: https://cursor.com/automations/f47c065b-b292-11f1-a3d8-362438fd9788
Sometimes - I can sometimes use Cursor
Hey, thanks for the reports. This isn’t anything in your setup, it’s an issue on our side.
Since early this morning (UTC), some Cloud Agents have been failing to start on Grok 4.7 with the error AI Model Not Found or The model you chose is not available when starting from cursor.com, the desktop app, or the API. That’s why the first agent ran, but the next one didn’t. The same issue is currently affecting Gemini 3.8 Flash in Cloud Agents as well. I’ve already passed this on to the team.
While we’re working on a fix, here are the current workarounds:
Once this is fully restored, Grok 4.7 in Cloud Agents will work again without you needing to do anything. We’ll reply in this thread when I have an update.
Cursor IDE
When I try to start a new cloud chat with Grok 4.7, the chat starts, then disappears, and the model selection reverts to the model I previously used.
The same behavior occurs when I try to use Grok 4.7 in an existing chat. Inference stops, and the following message appears:
The model you chose is not available
Choose a different model from the picker and try again.
This only happens with cloud chats. For local chats, I can use Grok 4.7.
MacOS
Version: 3.21.16 (Universal)
VS Code Extension API: 1.128.0
Commit: 8ae78e8eee1e63479c7e0504b664bc0a80c68000
Date: 2026-09-19T06:43:22.276Z
Layout: Agent Window
Build Type: Stable
Release Track: Default
Electron: 42.10.0
Chromium: 148.0.7778.280
Node.js: 24.18.1
V8: 14.8.178.38-electron.0
xterm.js: 6.1.0-beta.291
OS: Darwin arm64 27.0.0
Grok 4.7 / Extra High / Fast / 500k
Cannot since it disappears
No - Cursor works, but with this issue
Cursor IDE
when you want to use Grok 4.7 in projects it doesn’t let you.
Easy go to projects and use the new grok 4.7
it should work because it does work at workspaces in cursor but not at projects
Windows 10/11
Version: 3.21.16 (user setup)
VS Code Extension API: 1.128.0
Commit: 8ae78e8eee1e63479c7e0504b664bc0a80c68000
Date: 2026-09-19T06:43:22.276Z
Layout: Agent Window
Build Type: Stable
Release Track: Default
Electron: 42.10.0
Chromium: 148.0.7778.280
Node.js: 24.18.1
V8: 14.8.178.38-electron.0
xterm.js: 6.1.0-beta.291
OS: Windows_NT x64 10.0.19045
grok 4.7
No - Cursor works, but with this issue
Same here.
Same
Hey all,
Sorry about this! We’ve implemented a fix that should finish rolling out imminently (I just successfully started a Cloud Agent with Grok 4.7)
I can confirm that our automation triggered agents, are now starting as expected also with grok 4.7 chosen.
Thank you for your attention to this matter, quick resolve.
I also had an issue where the agent flips the current model from Grok 4.7 xhigh 256k to Grok 4.7 low 500k without me asking for it in cloud agents since early this morning. Is the fix related to this as well ? @Colin Thanks
