NVIDIA's ENPIRE Framework Enables AI Coding Agents to Autonomously Train Robots on Complex Tasks

According to NVIDIA research published on June 16, the NVIDIA GEAR lab—in collaboration with Carnegie Mellon University and UC Berkeley—unveiled ENPIRE, an agentic harness framework that enables AI coding agents to autonomously direct robot training. The framework allows AI agents to independently develop and refine training approaches for robots, with successful demonstrations including cutting zip ties and inserting GPUs into motherboards. ENPIRE operates through four modules that handle automatic task reset, policy refinement, parallel robot evaluation, and failure analysis. The team tested the framework with three AI coding agents: OpenAI's Codex with GPT-5.5, Anthropic's Claude Opus 4.7, and Moonshot AI's Kimi K2.6, each developing distinct algorithmic approaches through cycles of self-directed testing.
Disclaimer: The information on this page may come from third-party sources and is for reference only. It does not represent the views or opinions of Gate and does not constitute any financial, investment, or legal advice. Virtual asset trading involves high risk. Please do not rely solely on the information on this page when making decisions. For details, see the Disclaimer.
Comment
0/400
No comments