Kinematics Lab/OpenVLA grasping

VLAHardware: MaxRoadmap scaffold

OpenVLA grasping

Open-source VLA model for manipulation — no proprietary checkpoints.

An OpenKinematics engineer running robot demos from a live control console — OpenVLA grasping runs against the same observable edge stack used for testing and deployment.

This entry is a roadmap scaffold, not a finished demo. Its description records intended scope; the linked repository is the source of truth for current code.

About this demo

OpenVLA (7B parameter open Vision-Language-Action model) running on Jetson Thor-series modules for general-purpose grasping. Fine-tune on your own teleop data. Inspired by Jetson AI Lab Research Group's OpenVLA review.

Highlights

→Open weights (no proprietary lock-in)
→Fine-tunable on consumer GPU
→TensorRT-optimized export
→Drop-in replacement for proprietary VLAs

Supported robots

Any 6-DoF arm

Related demos

GR00T VLA pick-and-place

Language-conditioned manipulation with NVIDIA GR00T N1.5.

Open demo →VLA

VLM agent in Isaac Sim

Vision-language reasoning over a simulated robotics scene.

Open demo →Teleop

CockPit

Predefined teleop dashboard with dual cameras, 3D model, map, and controls.

Cookie Settings

We use cookies to analyse site traffic and personalise content. Read our Cookie Policy for details.