Nvidia Releases Open-Source ASPIRE Robotics System

Nvidia’s GEAR lab has made its ASPIRE v1.0 platform available to the public via GitHub.

Updated on Oct. 16, 2026 in Robotics

Isometric editorial illustration of a robotic arm in a lab, representing a framework for generalist robotics development. — Nvidia’s GEAR lab released its open-source ASPIRE 1.0 platform, providing a framework for generalist robotics to learn and execute tasks. AI Illustration. Upload story photo >

Live Poll

Do you believe rapid advancements in AI robotics will improve your daily life in the future?

Nvidia’s GEAR lab has released version 1.0 of the ASPIRE system, an open-source framework designed to provide an operating layer for generalist robotics. The technology uses large language models to write and refine Python control code for robots, enabling them to solve complex tasks.

Why it matters

The system provides an accessible platform for startups and external research labs to develop generalist robotics capabilities. By enabling robots to learn from observed performance, the tool aims to standardize the development of autonomous robotic control systems.

The ASPIRE system achieved a 99% success rate across 80 LIBERO-Pro benchmark tasks and a 72% success rate on Robosuite bimanual handover challenges. Performance varied significantly by language model, with success rates jumping from 15% to 99% when switching from GPT-6 Luna to Claude Opus 5.5.

The players

Nvidia GEAR Lab

This division operates as the research arm for Nvidia, focusing on advancing embodied artificial intelligence and robotic systems.

GitHub

This is a global software development platform that hosts millions of open-source projects and code repositories.

The details

The platform functions by using a large language model to write, execute, and iteratively rewrite Python control code based on real-time robot performance. Validated code fixes are then stored in a library to ensure the robot does not need to restart tasks from zero.

June 30, 2026: The ASPIRE research paper was originally posted to arXiv.
October 2026: Nvidia released version 1.0 of the ASPIRE system on GitHub.

The Tech Race

This release marks a shift toward open-source frameworks in the race to build autonomous generalist robots. By standardizing the control layer, the system moves the industry away from custom, proprietary code bases and toward a shared architecture for robot intelligence.

Software developers and robotics researchers can now utilize the Apache 2.0 licensed code to build and test their own robotic applications without building a control system from scratch. This could accelerate the development of more capable household and industrial robots by simplifying the integration of AI models.

The takeaway

Nvidia’s decision to open-source the ASPIRE platform highlights the growing importance of standardizing robot control software. By leveraging language models to write code, this framework offers a template for how future robots could learn and improve their performance through iterative execution.