Acheron Labs
"Build context-aware, reasoning autonomous robots that remember and learn."
Vienna-based robotics software startup building an LLM-driven control layer that lets Unitree's G1 humanoid and Go2 quadruped follow spoken commands.
Acheron Labs is an early-stage robotics software company headquartered in Vienna, Austria, with additional presence in Prague and Bratislava. The company builds a control and orchestration layer that connects large language models to physical robot hardware, letting operators issue natural-language instructions that the software converts into planned, executable robot behaviors. Rather than manufacturing robot bodies, Acheron focuses on the software stack: spatial reasoning, task planning, and integration with external apps and services in a workflow style comparable to automation tools such as n8n, plus fleet management and performance analytics for deployed robots.
The platform is built around ROS2 and currently supports a small set of commercially available robot platforms, including Unitree's G1 humanoid for navigation and manipulation tasks, the OpenARM robotic arm for vision-guided manipulation, and Unitree's Go2 quadruped for navigation and voice-driven control. In July 2026, Acheron Labs' toolkit -- referred to in coverage of the collaboration as "RobotKit" -- was demonstrated alongside LimX Dynamics' TRON 1 biped robot to show autonomous household operation, positioning the pairing as an example of an "open embodied AI ecosystem" in which a hardware maker's platform is paired with a third-party software layer for rapid application development.
Acheron's stated goal is to build context-aware, reasoning autonomous robots that remember and learn, combining large language and vision models so a robot can retain context across interactions rather than executing single, stateless commands. The company markets an SDK for developers to connect their own external software and hardware into the control loop, and it targets researchers and robotics developers building embodied-AI applications rather than end consumers.
As a young company -- public materials give no confirmed founding date beyond a 2025 site copyright notice -- Acheron Labs has not disclosed employee headcount, funding, or revenue. It maintains a presence on X (@acheron_labs), Instagram, and GitHub, and lists NVIDIA, OpenAI, Unitree, and Pollen Robotics among the platforms and partners referenced on its site. Its core differentiation versus other robotics middleware vendors is the explicit focus on LLM-driven natural-language control combined with cross-vendor hardware support spanning humanoids, robotic arms, and quadrupeds, rather than being tied to a single robot manufacturer's proprietary stack.
Primary type & automation activities this supplier delivers:
Acheron
Acheron is a reasoning AI platform adding natural-language control, spatial memory and fleet management to ROS2 robots such as Unitree G1 and GO2.
Contact Acheron Labs
WEBSITE
https://www.acheronlabs.comHEADQUARTERS
Austria
Company Facts
Founded
-
Primary Role
Software/Algorithm
Company Size
-
Primary Region
Europe
Annual Sales
-
Funding Stage
-
Funding Total
-
LimX Dynamics Demonstrates Autonomous Home Operations with TRON 1 and Acheron Labs' RobotKit
On July 13, 2026, LimX Dynamics showcased a customer case utilizing their open modular robot platform, TRON 1, to demonstrate autonomous home operation workflows with Acheron Labs' RobotKit. This collaboration illustrates the rapid development and deployment of embodied AI applications by combining RobotKit's development workflow with TRON 1's modular structure. This initiative is significant as it embodies LimX Dynamics' commitment to fostering an open embodied AI ecosystem that accelerates innovation among developers, researchers, and ecosystem partners. The demonstration serves as a practical example of development methods in the fields of embodied AI and physical AI, highlighting the potential for transforming operational tasks in various industries. Looking ahead, the focus will be on how robots can be integrated into real-world applications, particularly in high-risk environments. No further timeline was disclosed at the time of publication.
