Wall-OSS-0.5
Wall-OSS-0.5 is a 4B vision-language-action model from X Square Robot, combining a 3B vision-language backbone with action-generation components. It uses gradient-bridged co-training and continuous flow matching for robot actions.
Access and availability
- Code
- Public access
- Model files
- Public access
Training / fine-tuning code
- Access
- Public access
Verified on · 2026-10-30
- Code license
- Apache-2.0
Model files
| File | Access |
|---|---|
| x-square-robot/wall-oss-0.5 | Access |
At a glance
- Weight parameters
- 4.1622 B
- Released
- May 28, 2026
Model card
Wall-OSS-0.5 is a 4B vision-language-action model from X Square Robot, combining a 3B vision-language backbone with action-generation components. It uses gradient-bridged co-training and continuous flow matching for robot actions. https://arxiv.org/abs/2605.30877 https://huggingface.co/x-square-robot/wall-oss-0.5 https://github.com/X-Square-Robot/wall-x
Before you use this version: Available weights or code do not establish plug-and-play robot compatibility. Check the required configuration, data and interfaces in the source documentation.