NVIDIA just dropped NitroGen, an open vision-action model that learns to play games straight from pixels and controller inputs - trained on 40K hours of gameplay across 1,000+ titles. The fact that they're releasing the dataset and a universal simulator alongside it makes this genuinely useful for the research community, not just a flashy demo. Curious to see how quickly this gets adapted for robotics applications given the visual-to-action pipeline.
NVIDIA just dropped NitroGen, an open vision-action model that learns to play games straight from pixels and controller inputs - trained on 40K hours of gameplay across 1,000+ titles. 🎼 The fact that they're releasing the dataset and a universal simulator alongside it makes this genuinely useful for the research community, not just a flashy demo. Curious to see how quickly this gets adapted for robotics applications given the visual-to-action pipeline.
WWW.MARKTECHPOST.COM
NVIDIA AI Researchers Release NitroGen: An Open Vision Action Foundation Model For Generalist Gaming Agents
NVIDIA AI research team released NitroGen, an open vision action foundation model for generalist gaming agents that learns to play commercial games directly from pixels and gamepad actions using internet video at scale. NitroGen is trained on 40,000 hours of gameplay across more than 1,000 games and comes with an open dataset, a universal simulator, […] The post NVIDIA AI Researchers Release NitroGen: An Open Vision Action Foundation Model For Generalist Gaming Agents appeared first on Mar
0 Kommentare 1 Geteilt 66 Ansichten
Zubnet https://www.zubnet.com