{"slug": "nvidia-open-sources-curobov2-fast-a-robot-planner-that-works-in-4-4-ms", "title": "NVIDIA open-sources cuRoboV2-Fast, a robot planner that works in 4.4 ms", "summary": "NVIDIA researchers Balakumar Sundaralingam, Adithyavairavan Murali and Stan Birchfield have open-sourced cuRoboV2-Fast under Apache 2.0 through NVlabs, a robot motion planner that performs collision-free, torque-limited planning in 4.4 ms. Sundaralingam wrote on X on November 5th that the first cuRobo prototype in 2022 took 100 ms and consumed the whole GPU, making the new planner roughly 20 times faster after four years of kernel work. The release gives research teams and robotics companies a planner they can inspect and benchmark on their own arms while respecting payload and torque limits.", "body_md": "# NVIDIA open-sources cuRoboV2-Fast, a robot planner that works in 4.4 ms\n\n**Sundaralingam, who built the first 100 ms cuRobo prototype in 2022, is releasing NVIDIA's 4.4 ms planner with Adithyavairavan Murali and Stan Birchfield.**\n\n        By [Ryan Merket](https://runtimewire.com/author/ryan-merket)\n        · Published \n\nPrimary source: [Aligned News — AI Intelligence](https://x.com/balakumar_/status/2118408305857228876)\n\n## Why it matters\n\nMotion planning sits between a robot seeing an object and moving to grasp it. Cutting that step to 4.4 ms on stated hardware, while respecting payload and torque limits, reduces the time spent planning a path. Releasing it under Apache 2.0 through NVlabs gives research teams and robotics companies a planner they can inspect, benchmark on their own arms, and build products around.\n\n[Balakumar Sundaralingam (@balakumar_)](https://x.com/balakumar_?ref=runtimewire) started cuRobo with a planner that took 100 ms. \"In 2022, our first cuRobo prototype took 100 ms and consumed the whole GPU,\" he wrote [in a thread on X](https://x.com/balakumar_/status/2118408305857228876?ref=runtimewire) on November 5th. Four years of kernel work later, he is open-sourcing cuRoboV2-Fast under Apache 2.0, with collision-free, torque-limited motion planning in 4.4 ms.\n\nWhen you purchase through links in our articles, [we may earn a small commission](https://runtimewire.com/affiliate-monetization-standards). This doesn’t affect our editorial independence.", "url": "https://wpnews.pro/news/nvidia-open-sources-curobov2-fast-a-robot-planner-that-works-in-4-4-ms", "canonical_source": "https://runtimewire.com/article/nvidia-curobov2-fast-open-source-4-4-ms-robot-planner", "published_at": "2026-11-05 19:03:29+00:00", "updated_at": "2026-11-05 22:21:46.646960+00:00", "lang": "en", "topics": ["robotics", "ai-research", "ai-tools", "developer-tools"], "entities": ["NVIDIA", "cuRoboV2-Fast", "cuRobo", "Balakumar Sundaralingam", "Adithyavairavan Murali", "Stan Birchfield", "NVlabs", "Apache 2.0"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/nvidia-open-sources-curobov2-fast-a-robot-planner-that-works-in-4-4-ms", "markdown": "https://wpnews.pro/news/nvidia-open-sources-curobov2-fast-a-robot-planner-that-works-in-4-4-ms.md", "text": "https://wpnews.pro/news/nvidia-open-sources-curobov2-fast-a-robot-planner-that-works-in-4-4-ms.txt", "jsonld": "https://wpnews.pro/news/nvidia-open-sources-curobov2-fast-a-robot-planner-that-works-in-4-4-ms.jsonld"}}