--- format: "aidr-story-markdown/v1" id: "5b96dde6fe39bf244d6e228bfa8b54fa16e6670ceea9fdfb84907e3aedea1868" canonical_url: "https://aidr.today/5b96dde6?lang=en" title: "Verne Robotics' V0.5 Robot Model Can Learn New Industrial Tasks in Under 2 Hours" lang: "en" requested_lang: "en" available_langs: ["en","vi"] translation_fallback: null fallback_fields: [] published_at: "2026-11-05T19:08:49.000Z" category: "Models" topics: ["agent"] source_urls: ["https://marketbrief.now/ai/verne-robotics-v05-robot-model-can-learn-new-industrial-tasks-in-under-2-44dde0a9","https://huggingnews.com/ai/verne-robotics-v05-robot-model-can-learn-new-industrial-tasks-in-under-2-44dde0a9"] summary: "Verne Robotics' V0.5 robot model can learn new industrial tasks in under 2 hours. V0.5 was trained on more than 1.7M demonstrations using the Wan2.2 video generator. When tested on the RoboTwin 2.0 benchmark, V0.5 achieved a success rate of 73.4% against randomized clutter and lighting. The model runs inference directly on the robot hardware to remove the need for constant cloud connectivity." --- # Verne Robotics' V0\.5 Robot Model Can Learn New Industrial Tasks in Under 2 Hours > [Open the canonical story]() **Published:** 2026-11-05T19:08:49.000Z **Category:** Models **Topics:** agent ## Summary Verne Robotics' V0\.5 robot model can learn new industrial tasks in under 2 hours\. V0\.5 was trained on more than 1\.7M demonstrations using the Wan2\.2 video generator\. When tested on the RoboTwin 2\.0 benchmark, V0\.5 achieved a success rate of 73\.4% against randomized clutter and lighting\. The model runs inference directly on the robot hardware to remove the need for constant cloud connectivity\. ## Sources - [Story source]() - [Story source]()