Next robot →

ALOHA 2 / Trossen AI Stationary

Stanford & Google DeepMind (design), Trossen Robotics (kits, USA) · 2023 · All Stanford University robots · All Trossen Robotics robots

$23,995.95
AvailablePython from a laptopBimanual teleopImitation learningOpen-source hardwareLeRobot / OpenPi
Weeks+Shipping: 2–3 weeks production.
Remote control: Everything runs from a Linux PC or laptop: ROS 2 and Python scripts record leader-arm demonstrations and run trained policies on the follower arms.

Overview

ALOHA (Stanford, 2023) showed that low-cost leader/follower arms plus imitation learning could thread zip ties and handle batteries. ALOHA 2 (Google DeepMind, 2024) improved grippers, ergonomics and robustness and open-sourced the design. Trossen sold the official kits, then rebranded the line as Trossen AI with its own WidowX AI arms. It is the reference setup for bimanual robot-learning research and many public datasets.

At a glance
Type
Robot arm · Research & collaborative arms
Maker
Stanford & Google DeepMind (design), Trossen Robotics (kits, USA)
Released
2023
Price
$23,995.95
Status
Available
Best for
Professionals & labs
Programming
Open source (hardware + software)

Price and shipping

Current Trossen Stationary AI kit (4 WidowX AI arms, 4 RealSense D405 cameras, frame). The original ALOHA was built on a ~$20k budget. Trossen's ALOHA Stationary kits are now discontinued.

Shipping: 2–3 weeks production. Ordered from Trossen Robotics in the US. Average production time is 2–3 weeks before shipping. Table not included. Trossen ↗

Specifications

CategoryRobot arms · Research & collaborative arms
Released2023
CountryUSA
AvailabilityAvailable
ProgrammabilityOpen source (hardware + software)
Built-in autonomyNone / scripted · Moves only when controlled, or plays preset motions
LLM supportOfficial LLM tools · Trossen maintains an OpenPI fork and tutorials for running the language-conditioned pi0 and pi0.5 VLA models on its kits. Trossen Robotics blog (OpenPI integration) ↗
MuJoCo simulationOfficial model · In MuJoCo Menagerie (aloha), Trossen also publishes Stationary AI MJCF. MuJoCo Menagerie ↗
NVIDIA Isaac simulationOfficial model · Trossen's trossen_ai_isaac provides Stationary AI USD models and Isaac Lab tasks. trossen_ai_isaac ↗
Gazebo simulationMaker URDF only · Trossen publishes arm URDFs, no Gazebo launch files found. trossen_arm_description ↗
Best forProfessionals & labs
Vision sensorsCamera, Depth camera
ActuatorsSmart servos (Dynamixel, Feetech…)
AudioNo audio
Compute typeLaptop / PC as the brain
Remote controlPython from a laptop
DOF4 arms × 6 DOF + grippers (2 leader, 2 follower)
Reach700 mm per WidowX AI arm (Stationary AI), ViperX 300 S followers on ALOHA 2
Payload1.5 kg per WidowX AI arm
Repeatability1 mm
Weight4 kg per WidowX AI arm
ControllerHost Linux PC/laptop (preloaded laptop with ALOHA kits), iNerve arm controllers on Trossen AI
Sensors4× Intel RealSense D405 cameras (Stationary AI), joint feedback, gravity compensation on leader arms (ALOHA 2)
SDKALOHA codebase (ROS 2), MuJoCo model, Trossen arm SDK, Hugging Face LeRobot, OpenPi, Isaac, Gazebo

Common questions

How much does ALOHA 2 / Trossen AI Stationary cost?

ALOHA 2 / Trossen AI Stationary is listed at $23,995.95. Current Trossen Stationary AI kit (4 WidowX AI arms, 4 RealSense D405 cameras, frame). The original ALOHA was built on a ~$20k budget. Trossen's ALOHA Stationary kits are now discontinued.

Can ALOHA 2 / Trossen AI Stationary be bought today?

Yes, it is on sale. Typical shipping time is 2–3 weeks production. Ordered from Trossen Robotics in the US. Average production time is 2–3 weeks before shipping. Table not included.

Who makes ALOHA 2 / Trossen AI Stationary?

ALOHA 2 / Trossen AI Stationary is made by Stanford & Google DeepMind (design), Trossen Robotics (kits, USA). It came out in 2023.

Can ALOHA 2 / Trossen AI Stationary be programmed?

Open source (hardware + software). ALOHA codebase (ROS 2), MuJoCo model, Trossen arm SDK, Hugging Face LeRobot, OpenPi, Isaac, Gazebo.

Does ALOHA 2 / Trossen AI Stationary have AI chat or LLM support?

Official LLM tools. Trossen maintains an OpenPI fork and tutorials for running the language-conditioned pi0 and pi0.5 VLA models on its kits.

Ratings & reviews

Related guides

Similar robots

Sources

ALOHA 2 project ↗ · Original ALOHA ↗

Image credits