🔥Telegram İfşa

Excited that we are launching Gemini Robotics 1.5, a pair of models that can work together to solve long horizon tasks. Gemini Robotics-ER 1.5, the high-level reasoning brain, is available for public preview in AIStudio and Gemini API for robotics developers around the world:

#Gemini Robotics is spreading at #CoRL2025. 🔥 Check out a finetuned on-device version of our model at the booth of FRANKA ROBOTICS nearby. And if you’d like to test our models and tools on your platform, consider signing up to our trusted tester program: docs.google.com/forms/d/1sM5Gq…

Had so much fun at CoRL 2025! If only I had some robots to pack my bag for my way home... Gemini Robotics 1.5 finetuned with some additional data on this scene:

Even though, #CoRL2025 is already over, there are still a few more results we are excited to share. 🤩 #Gemini Robotics 1.5 is a first glimpse into the era of agentic robotics. Here’s an example of how a robot kitchen assistant could look like one day. 👇

💯 When I saw for the first time how the Aloha picked tools from Franka’s backpanel and the Franka opened the pear sorter, I could genuinely feel that this was a step-change in VLA training. Bimanual manipulation across all platforms feels very attainable now. 🚀

I remember how many discussions and attempts we’ve had in the team in the past about solving NIST boards. And then, #Gemini Robotics 1.5 happened and it… just… worked. 🤯 VLAs with agentic orchestration are a powerful combo which, IMO, will still yield many more surprises. 🚀

A pity that our demo at #CoRL2025 had to fight with the conference centre’s WiFi. But when latency is not an issue, our #Gemini Robotics 1.5 model also shows impressive dexterity during precise insertion and can even close a ziplock bag. 🦾

When I had just started at DeepMind, I was awestruck learning about the fact that a Tokamak fusion reactor could be controlled by an RL-tuned network. Fast forward four years and I’m even more excited about where this is all heading. 🤩🚀

The mission is calling! #Google DeepMind Robotics is looking for talented researchers to join our team and help build the new frontier of robotics intelligence through world modeling. job-boards.greenhouse.io/deepmind/jobs/…

We're starting to hire for our 2026 Olmo interns! Looking for excellent students to do research to help build our best models (primarily enrolled in Ph.D. with experience or interest in any area of the language modeling pipeline).

Our Open-Endedness team at Google DeepMind is hiring student researchers. Come work on some cool research with the super talented Cong Lu 🤖🧪 🚀

Come work with the super talented Mikayel Samvelyan as a student researcher in our Open-Endedness team at Google DeepMind 🤖 ♾ 🚀

Nathan’s blog posts on RLHF have always been an absolute treasure trove for me who, as a robotics researcher, tried to stay on top of what’s happening in the RLHF space on LLMs. So great to see that bundled wisdom now in printed form, too. 🤓 Awesome cover, btw! 🍒🍰

Excellent take on how to better understand the “jagged frontier” of AI’s economical impact. I agree: if the outcome of a task is easily verifiable, it can (and will) be learned and done by AIs. Very neat analogy, especially when you’re working on RL. 🤓