Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: c7b0f97b95
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 1 potential issue.
Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable autofix in the Cursor dashboard.
JannikSt
left a comment
JannikSt
left a comment
There was a problem hiding this comment.
As mentioned in my comment - should clearly mark gpu count as closed beta atm
packages/prime-sandboxes/README.md
Outdated
| sandbox_client.delete(sandbox.id) | ||
| ``` | ||
|
|
||
| ### GPU Sandbox Example |
There was a problem hiding this comment.
should flag that this is a closed beta right now
2 participants
The renaming of STOPPED to PAUSED aligns with the backend name change. This is vestigial code not yet used so, technically breaking, but ok to do.
Note
Medium Risk
Introduces a new Connect-RPC execution path (with generated protobufs and new deps) and changes CLI create semantics, which could affect command execution behavior and error handling for GPU vs CPU sandboxes.
Overview
Adds GPU sandbox support end-to-end: models now include
gpu_type,CreateSandboxRequestvalidatesgpu_typewhengpu_count>0, and sandbox statusSTOPPEDis renamed toPAUSED.Updates command execution to auto-select transport (REST for CPU, Connect RPC streaming for GPU) with new generated protobuf modules and
connect-python/protobufdeps, plus improved mapping of gateway502sandbox_not_foundinto a not-running error.CLI updates include
prime sandbox createGPU flags (--gpu-count/--gpu-type) with enforced argument rules (no docker image for GPU; internal placeholder image), clearer GPU display in list/get, and explicit guards that block port exposure and SSH for GPU sandboxes; docs and tests are updated accordingly.Written by Cursor Bugbot for commit cf84c59. This will update automatically on new commits. Configure here.