Open Source RoboticsRobots, software & learning

Wall-X

Software

VLA training/inference source for WALL-series robot policies, with separate published checkpoint listings and CUDA operator builds. [1] [3]

Installation & first result →Official websiteGetting started

For
Robot-learning researchers working with CUDA and LeRobot data [1]
Source rights
Some licenses are unresolved. Review licenses
Unknowns
operating system, compute requirements not fully established. Review gaps

Source researched · Not tested by us
Sources checked · Research details

Licenses & related hardware
License
Apache-2.0 code; checkpoint rights separate · license not fully assessed for model weights · Details by part

Requirements · Research details and dates

On this page

Overview

Wall-X contains training and inference implementation for WALL-series robot policies, including data preparation, action branches and serving utilities. Treat a named checkpoint and its action configuration as the unit of selection. A repository containing several policy families does not establish that one installation or normalization setting is interchangeable across them. [1] [3]

Is this for you?

Known limitations
  • WALL-OSS0.5 card and repository Python environment instructions differ.
  • Selected checkpoint is public/ungated in metadata, but no checkpoint license declaration was identified in card/metadata; code Apache2 does not settle weight rights.
  • No CUDA build, checkpoint payload, inference, closed-loop control or data rights validated.

Editorial scope assessment from the cited implementation/documentation; no execution or compatibility test.

[1] [2] [3] [4] [30] · Editorial assessment

What you need

Dependencies
  • Repository route uses Python3.10 while selected WALL-OSS0.5 model card uses Python3.12
  • Requirements, flash-attn2.8.3, DMuon and separately installed LeRobot are documented
  • CUDA operator compilation during editable install requires a matching PyTorch/CUDA build environment

[1] [2] [3] [4] [30]

Installation & first result

Open the upstream getting-started guide [1]

  • Select WALL-OSS0.5 or another named checkpoint and its matching norm_key/action setup; reconcile model-card and code-repository environment instructions before attempting installation.
  • Preserve exact source/dependency/model versions and evaluate the selected workflow before connecting physical hardware or reusing data.

Editorial scope assessment from the cited implementation/documentation; no execution or compatibility test.

[1] [2] [3] [4] [30] · Editorial assessment

Installation

Select WALL-OSS0.5 or another named checkpoint and its matching norm_key/action setup; reconcile model-card and code-repository environment instructions before attempting installation.

Editorial scope assessment from the cited implementation/documentation; no execution or compatibility test.

[1] [2] [3] [4] [30] · Editorial assessment

Hardware, models & data

Inputs and outputs

Robot image/state/language inputs to configured policy action predictions; training and serving source provided.

Editorial scope assessment from the cited implementation/documentation; no execution or compatibility test.

[1] [3] · Editorial assessment

Project notes

Resolve two documented installation routes

The selected repository README uses a Python 3.10 environment, while the WALL-OSS-0.5 model card shows Python 3.12. Reconcile the intended checkpoint and environment rather than silently combining those instructions. The repository documents requirements installation, flash-attn 2.8.3, DMuon and a separate LeRobot installation. Its editable-install step builds included CUDA operators using the existing PyTorch/CUDA environment. No CUDA compiler or package compatibility was tested here. [1] [30]

The repository requires --no-build-isolation for its editable Wall-X install so the build can use the installed PyTorch package. Its separate LeRobot step uses --no-deps to avoid overriding the requirements environment. DMuon is documented for the default training configurations; its presence is not evidence of a measured inference requirement. Preserve those distinctions while resolving the version set. [1]

The WALL-OSS-0.5 card associates robot embodiments with norm_key settings. Preserve the relevant robot state/action dimensions and normalization key before evaluating predictions. The repository's fake_inference script and open-loop plotting utilities are investigation tools, not evidence of a closed-loop physical deployment. No inference or command generation was run by this catalog. [1] [30]

What we checked

Source research only. Selected implementation/model text, repository tree and license declarations were inspected. No upstream compilation, dependency installation, simulator, network/USB/device integration, data collection, benchmark or physical robot test was performed. Source files are linked rather than redistributed; external services and paid products are not endorsed.

Not yet established

  • Requirements: operating system, host computer
  • Version-specific hardware compatibility (ecosystem tags alone do not establish it)

Updates

Upstream last updated
Not established
Latest release
Not established
Profile updated
Nov 1, 2026
Sources last checked
Nov 1, 2026
Next review
Dec 1, 2026
Reviewed revision
ce2a1ae

Source files and licenses

Distinct robot-facing model contribution with actual implementation/model content and a bounded documented next step. Review covers source artifacts, not operational performance or a complete tested deployment.

LayerAvailability / license
softwareavailable · Apache-2.0 code; checkpoint rights separate
Declaration for the scoped source implementation/documentation; dependencies, external services, model/data assets and inherited file-specific terms require separate checks. No blanket redistribution or operational clearance. [1] [2] [3]
model weightsavailable · License not fully assessed
Public ungated checkpoint listing and model card inspected at metadata revision b7dc23a82dbf70a65859c9fdd596703234597028; weight binaries not downloaded or executed. This does not settle backbone/data or other checkpoint terms. [30]
documentationavailable · Apache-2.0 code; checkpoint rights separate
Declaration for the scoped source implementation/documentation; dependencies, external services, model/data assets and inherited file-specific terms require separate checks. No blanket redistribution or operational clearance. [1] [2] [3]

This profile is scoped to source implementation/model descriptions and an initial inspection workflow; no physical manufacturing design, pretrained checkpoint or dataset is selected for redistribution or tested deployment.

Code availability does not settle checkpoint rights

The code repository declares Apache-2.0. The selected checkpoint's card and public model metadata identify a published, ungated listing at metadata revision b7dc23a82dbf70a65859c9fdd596703234597028, but no checkpoint license declaration was identified in that card/metadata scope. Do not automatically transfer the code license to the weights or training data. [2] [30]

This is source and model-card research. Weight binaries, training data, GPU-memory requirements, a functioning server, policy quality and robot behavior were not independently validated.

References

Each source links to the exact revision we read. “Used for” shows where this page relies on it.

Correction guidance · Discussion is not enabled · Profile format 0.7.0-draft.5

Change history