bugfix: pip_install_packages="" now doesn't break sandbox#633
Merged
mikasenghaas merged 1 commit intomainfrom Dec 14, 2025
Merged
Conversation
mikasenghaas
approved these changes
Dec 14, 2025
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
In
vf.PythonEnv, thepip_install_packagesargument causes Sandboxes to fail. Due to the error catching, this is only visible in the tool-outputs. The model still continues working (and succeeds in many environments like math-python).The error is caused by the
_START_COMMAND_TEMPLATEhavingpip install -q {pip_install_packages}in it, which errors whenpip_install_packages = "". This is now replaced with{pip_install_command}, which in turn is guaranteed to be empty if there are no packages to install.Previously, running
uv run vf-eval math-python -n 1 -r 1 -m gpt-5-mini -a '{"pip_install_packages": ""}'resulted in outputs like this, which resulted in full reward and a functional rollout, just no usable sandboxes:Now, the sandboxes work (though the models still assume the presence of numpy, scipy, and sympy for math-python; but that's not an issue with the sandboxes):
Type of Change
Testing
uv run pytestlocally.Checklist
Additional Notes