DeepMind Unveils Gemini Robotics 1.5, ER 1.5

On Sept. 25, 2025 Google DeepMind unveiled two robotics models—Gemini Robotics 1.5 (a vision‑language‑action model) and Gemini Robotics‑ER 1.5 (an embodied‑reasoning model). The two‑model approach separates acting from reasoning so robots can plan multi‑step tasks, perform spatial and temporal reasoning, and call digital tools (for example, Google Search) to ground decisions. DeepMind says this design improves generalization across robot platforms and supports complex use cases such as sorting laundry by color or packing a suitcase using local data. Gemini Robotics‑ER 1.5 is available to developers via the Gemini API in Google AI Studio, while Gemini Robotics 1.5 is currently offered to select partners in preview. DeepMind positions these releases as a step toward agentic, simulated reasoning in physical systems.