feat: add experimental LeRobot policy runtime - #3315
Open
TomCC7 wants to merge 7 commits into
Open
Conversation
Codecov Report❌ Patch coverage is @@ Coverage Diff @@
## feat/galaxea-a1z-hardware #3315 +/- ##
=============================================================
+ Coverage 75.18% 75.23% +0.05%
=============================================================
Files 1145 1146 +1
Lines 110038 110632 +594
Branches 9900 9936 +36
=============================================================
+ Hits 82728 83237 +509
- Misses 24532 24544 +12
- Partials 2778 2851 +73
Flags with carried forward coverage won't be shown. Click here to find out more.
... and 14 files with indirect coverage changes 🚀 New features to boost your workflow:
|
TomCC7
commented
Jul 31, 2026
TomCC7
force-pushed
the
feat/experimental-lerobot-policy
branch
from
July 31, 2026 23:35
a8ab2d8 to
486d959
Compare
TomCC7
marked this pull request as ready for review
July 31, 2026 23:58
TomCC7
requested review from
Dreamsorcerer,
leshy,
mustafab0,
paul-nechifor and
spomichter
as code owners
July 31, 2026 23:58
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Contribution path
Problem
DimOS needs an experimental runtime for executing trained LeRobot policies against live robot observations without coupling generic learning utilities to A1Z hardware or promoting an unstable policy API into a core package.
LeRobot 0.6 also requires Python 3.12+, Transformers 5, Hugging Face Hub 1.x, and NumPy <2.3. Those requirements conflict with the existing DimOS perception and development environments, which use Transformers 4.53 and NumPy 2.3.
Solution
LeRobotPolicyModuleunderdimos.experimental.robot_policywith named, lazily loaded checkpoints, typed streams, RPC status/control, and background-skill support.How to Test
Install and smoke-test the actual runtime environment:
uv sync --extra lerobot --no-default-groups --frozen python -c "import importlib.metadata as m; import dimos.experimental.robot_policy.lerobot; print(m.version('lerobot'), m.version('transformers'))"Expected versions: LeRobot 0.6.0 and Transformers 5.5.x.
Run the policy tests and CI-equivalent type check:
Validated locally: 16 policy tests passed; mypy passed across 935 source files; changed-file pre-commit hooks and
uv lock --checkpassed.AI assistance
OpenAI Codex with GPT-5 implemented and validated the change under direct user guidance. The user reviewed and approved the scope, package placement, dependency strategy, and stacked-PR structure.
Checklist