InfoQ - Game Development - News https://www.infoq.com InfoQ Game Development News feed SIMA 2 Uses Gemini and Self-Improvement to Generalize across Unseen 3D and Photorealistic Worlds https://www.infoq.com/news/2025/12/sima-2-gemini-agent/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=Game+Development-news <img src="https://res.infoq.com/news/2025/12/sima-2-gemini-agent/en/headerimage/generatedHeaderImage-1766979483818.jpg"/><p>Google DeepMind researchers introduced SIMA 2 (Scalable Instructable Multiworld Agent), a generalist agent built on the Gemini foundation model that can understand and act across multiple 3D virtual game environments. The SIMA 2 architecture uses a Gemini Flash-Lite model trained on a mixture of gameplay and Gemini pretraining data.</p> <i>By Vinod Goje</i> Generative AI AIOps AI Architecture Game Development Google DeepMind AI, ML & Data Engineering news Mon, 29 Dec 2025 11:05:00 GMT https://www.infoq.com/news/2025/12/sima-2-gemini-agent/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=Game+Development-news Vinod Goje 2025-12-29T11:05:00Z /news/2025/12/sima-2-gemini-agent/en