Serkan Cabi (@serkancabi) 's Twitter Profile
Serkan Cabi

@serkancabi

ID: 76233998

calendar_today22-09-2009 03:03:56

2,2K Tweet

845 Followers

397 Following

Google DeepMind (@googledeepmind) 's Twitter Profile Photo

Introducing Veo: our most capable generative video model. 🎥 It can create high-quality, 1080p clips that can go beyond 60 seconds. From photorealism to surrealism and animation, it can tackle a range of cinematic styles. 🧵 #GoogleIO

Demis Hassabis (@demishassabis) 's Twitter Profile Photo

For a long time, we’ve been working towards a universal AI agent that can be truly helpful in everyday life. Today at #GoogleIO we showed off our latest progress towards this: Project Astra. Here’s a video of our prototype, captured in real time.

Google DeepMind (@googledeepmind) 's Twitter Profile Photo

Introducing 2️⃣ new AI systems for robotics: 🤖 ALOHA Unleashed to perform two-armed manipulation tasks 🦾 DemoStart to control a multi-fingered robotic hand They learned to tackle a range of actions requiring dexterity. Here's how. 🧵 dpmd.ai/3MI2mkT

Serkan Cabi (@serkancabi) 's Twitter Profile Photo

LLMs can now touch the world. Finally, we are sharing what we have been working on for a while. There is so much to explore about what Gemini Robotics can do. Check out the videos and the tech report. deepmind.google/discover/blog/…

Convergent Research (@convergent_fros) 's Twitter Profile Photo

we made a map! gap-map.org is a tool we built to help you explore the landscape of R&D gaps holding back science - and the bridge-scale fundamental development efforts that might allow humanity to solve them, across almost two dozen fields

we made a map!

gap-map.org is a tool we built to help you explore the landscape of R&D gaps holding back science - and the bridge-scale fundamental development efforts that might allow humanity to solve them, across almost two dozen fields
Oriol Vinyals (@oriolvinyalsml) 's Twitter Profile Photo

Hello Gemini 2.5 Flash-Lite! So fast, it codes *each screen* on the fly (Neural OS concept 👇). The frontier isn't always about large models and beating benchmarks. In this case, a super fast & good model can unlock drastic use cases. Read more: blog.google/products/gemin…

Google DeepMind (@googledeepmind) 's Twitter Profile Photo

We’re bringing powerful AI directly onto robots with Gemini Robotics On-Device. 🤖 It’s our first vision-language-action model to help make robots faster, highly efficient, and adaptable to new tasks and environments - without needing a constant internet connection. 🧵

Dhruv Shah (@shahdhruv_) 's Twitter Profile Photo

Yesterday, we live demo-ed a “generalist” VLA for (I think) the first time ever to a broad audience Robotics: Science and Systems. Bring any object. Ask anything. New environment, new instructions, no fine-tuning. Just impeccable vibes! ✨

Serkan Cabi (@serkancabi) 's Twitter Profile Photo

We helped making gravitational wave detection more sensitive. It was such a pleasure to be part of this project where physics meets machine learning. Google DeepMind is a wonderful and rare place where projects like this become possible.

Serkan Cabi (@serkancabi) 's Twitter Profile Photo

We just released our new generation of Gemini Robotics. It can control multiple robots, plan and execute more steps and even transfer skills from one robot to another.

Konstantinos Bousmalis (@bousmalis) 's Twitter Profile Photo

One of the coolest features in our newest release for Gemini Robotics 1.5 was using the wrist camera of one of the arms of our bi-arm Franka robot for active vision! Without this capability the model would not have enough visibility to solve some of our most dexterous tasks.

Ksenia Konyushkova (@ks_konyushkova) 's Twitter Profile Photo

Meet Gemini Robotics 1.5! Our ER model brings embodied reasoning to the real world. It understands high-level goals like "pack ingredients for mushroom risotto" with planning & success detection. Also, check out the cool "active vision" behavior – observing the action up close!🔍

Ksenia Konyushkova (@ks_konyushkova) 's Twitter Profile Photo

Interacting with robots just got easier with the agentic capabilities of Gemini Robotics 1.5. Talk to the robot or show it things! See how the ER model reads a handwritten list on paper and packs the tools for the job.

Konstantinos Bousmalis (@bousmalis) 's Twitter Profile Photo

You have to watch this! For years now, I've been looking for signs of nontrivial zero-shot transfer across seen embodiments. When I saw the Alohas unhang tools from a wall used only on our Frankas I knew we had it! Gemini Robotics 1.5 is the first VLA to achieve such transfer!!

Norman Di Palo (@normandipalo) 's Twitter Profile Photo

This is a remarkable result, let me explain. These tasks were only shown on a single robot embodiment. With the advancements of Gemini Robotics 1.5, *robots can learn from data from other robots* You taught a humanoid to pack gifts? Now a bi-arm Franka knows how to do it too

Konstantinos Bousmalis (@bousmalis) 's Twitter Profile Photo

Another agentic example and one of the demos we showed Conference on Robot Learning! Loved demonstrating the capabilities of Gemini Robotics 1.5 all week and it was fun to see how excited people were to interact with it in their own language with their own writing, drawings, and objects!

Caden Lu (@jyluxx) 's Twitter Profile Photo

Interacting with Gemini Robotics 1.5 is so fun! Our Embodied Reasoning model planned the multi-step task and orchestrated our Vision Language Action model for precise execution!

Google DeepMind (@googledeepmind) 's Twitter Profile Photo

We’re scaling up robotics in Europe. 🤖 Our Robotics Accelerator is tailored for startups and designed to bridge the gap between technology and business, powering the next generation of physical agents.