Loading video...

Video Failed to Load

Go Home

Gemini 3.0 Flash makes video understanding easier for robotics. Check out this demo where an Aloha robot executes complex tasks after a video is turned into a simple list of JSON action steps.

googleaidevs's profile picture

Google AI Developers

116,441 subscribers

47,167 views • 5 months ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Want to see Gemini 2.0 Flash Thinking in action? Check out this demo where the model solves a physics problem and explains its reasoning.
1:24

Want to see Gemini 2.0 Flash Thinking in action? Check out this demo where the model solves a physics problem and explains its reasoning.

Jeff Dean

105,766 views • 1 year ago

So dang enchanted with the Gemini 3.0 Flash support in Browserbase, especially for tedious data tasks! Check out the video below - Gemini navigates to the USGS website for seismic data, filters on 3D Multichannel Seismic datasets, adds all of the options for California, and then downloads as a CSV list. ❤️ You could even then have the data science agent in Colaboratory post-process and visualize the surveys!
1:22

So dang enchanted with the Gemini 3.0 Flash support in Browserbase, especially for tedious data tasks! Check out the video below - Gemini navigates to the USGS website for seismic data, filters on 3D Multichannel Seismic datasets, adds all of the options for California, and then downloads as a CSV list. ❤️ You could even then have the data science agent in Colaboratory post-process and visualize the surveys!

👩‍💻 Paige Bailey

25,681 views • 7 months ago

📒 Automating market research with Google AI Studio Gemini Search grounding + JSON mode this evening: 1⃣ Use Gemini 1.5 Flash + Search grounding to pull recent data about a list of companies, their product offerings, valuations, more. 2⃣ Use JSON mode with Gemini 1.5 Flash-8B to extract out specific details, for ~$0. 3⃣ Write data back to Google Sheets from Colaboratory:
0:36

📒 Automating market research with Google AI Studio Gemini Search grounding + JSON mode this evening: 1⃣ Use Gemini 1.5 Flash + Search grounding to pull recent data about a list of companies, their product offerings, valuations, more. 2⃣ Use JSON mode with Gemini 1.5 Flash-8B to extract out specific details, for ~$0. 3⃣ Write data back to Google Sheets from Colaboratory:

👩‍💻 Paige Bailey

43,383 views • 1 year ago

Combining real-time interactivity, task understanding, and full-body action prediction on a humanoid is so, so hard. Here's an example where we bring all of these together in Gemini Robotics 2 🤖🧠
1:07

Combining real-time interactivity, task understanding, and full-body action prediction on a humanoid is so, so hard. Here's an example where we bring all of these together in Gemini Robotics 2 🤖🧠

Annie Chen

40,103 views • 8 days ago

Haiku 4.5 has finally been dethroned by GPT 5.4 mini: - faster - cheaper - smarter - better at tool calling Gemini 3.0 flash still has a place in “turn this video/audio/blob of text into json”, but for any small model task that requires an agent loop (sub agents, search, quick fixes, browser use, etc.) GPT-5.4 mini is the new king
11:04

Haiku 4.5 has finally been dethroned by GPT 5.4 mini: - faster - cheaper - smarter - better at tool calling Gemini 3.0 flash still has a place in “turn this video/audio/blob of text into json”, but for any small model task that requires an agent loop (sub agents, search, quick fixes, browser use, etc.) GPT-5.4 mini is the new king

Ben Davis

67,624 views • 4 months ago

Following our Gemini 3.5 Flash launch at #GoogleIO, check out this demo of what it can do right inside the Gemini app. Watch 3.5 Flash build an interactive circuit helper, outputting a step-by-step physical build guide alongside a working simulation. It’s a great example of how the Gemini app can help students learn visually, breaking down new or complex concepts and guiding them through hands-on subjects interactively. 3.5 Flash is available globally today!
0:59

Following our Gemini 3.5 Flash launch at #GoogleIO, check out this demo of what it can do right inside the Gemini app. Watch 3.5 Flash build an interactive circuit helper, outputting a step-by-step physical build guide alongside a working simulation. It’s a great example of how the Gemini app can help students learn visually, breaking down new or complex concepts and guiding them through hands-on subjects interactively. 3.5 Flash is available globally today!

koray kavukcuoglu

44,144 views • 2 months ago

The latest episode of "What's New in Google Accessibility" is out now. Learn about new accessibility features including captions for Gemini Live, updates for Google Meet, and a Gemini-powered tool that simplifies complex text for easier understanding. #a11y #Accessibility #WhatsNewInGoogleAccessibility #Google
5:07

The latest episode of "What's New in Google Accessibility" is out now. Learn about new accessibility features including captions for Gemini Live, updates for Google Meet, and a Gemini-powered tool that simplifies complex text for easier understanding. #a11y #Accessibility #WhatsNewInGoogleAccessibility #Google

Google Accessibility

23,611 views • 9 months ago

Last month we announced Gemini Robotics (GR) and Gemini Robotics-ER (GR-ER). GR-ER is a powerful VLM specialised for spatial understanding, including detecting object poses in 2D/3D, pointing, and even *predicting grasp poses*. Take a look at this demo. Details below. 🧵
1:42

Last month we announced Gemini Robotics (GR) and Gemini Robotics-ER (GR-ER). GR-ER is a powerful VLM specialised for spatial understanding, including detecting object poses in 2D/3D, pointing, and even *predicting grasp poses*. Take a look at this demo. Details below. 🧵

Norman Di Palo

56,437 views • 1 year ago

Learn how to build a Gemini 3 Deep Agents with this video from LangChain. These agents can reason through complex, long horizon tasks by: > Breaking goals into actionable steps > Delegating specific work for specialized models > Leveraging file systems and code execution
7:49

Learn how to build a Gemini 3 Deep Agents with this video from LangChain. These agents can reason through complex, long horizon tasks by: > Breaking goals into actionable steps > Delegating specific work for specialized models > Leveraging file systems and code execution

Google AI Developers

19,484 views • 8 months ago

Humanoid robotics is entering a new phase： Xiaomi Robotics 0 (XR-0) explores learning manipulation directly from large-scale human videos — aligning video understanding with robot embodiment to generate executable actions. Watch humans->learn representations ->map to robot control. Robot data is scarce. Internet video is not. If video pretraining transfers reliably into embodied control, the scaling law of robotics changes. Deployment speed becomes a function of model alignment, not just hardware iteration. We’re moving from engineered skills to learned behavior. That’s a structural shift. project:
0:52

Humanoid robotics is entering a new phase： Xiaomi Robotics 0 (XR-0) explores learning manipulation directly from large-scale human videos — aligning video understanding with robot embodiment to generate executable actions. Watch humans->learn representations ->map to robot control. Robot data is scarce. Internet video is not. If video pretraining transfers reliably into embodied control, the scaling law of robotics changes. Deployment speed becomes a function of model alignment, not just hardware iteration. We’re moving from engineered skills to learned behavior. That’s a structural shift. project:

CyberRobo

18,356 views • 5 months ago

🧵 1/7 I gave a Flash model a written spec for a 3D physics game. It broke the task into steps, wrote the code, used the tools, ran the project, and eventually produced a billiards game I could actually play. Ant Ling Ling-3.0-flash has 124B total parameters, but activates just 5.1B per token. That low activation count didn’t limit it to a simple one-shot demo. It kept a multi-stage 3D build moving from spec to playable result. This test made one thing clear: stop using your biggest model for every step of an agent workflow. The project may be complex. Most of the execution steps inside it don’t need a heavyweight planner to rethink everything from scratch. Ling-3.0-flash is built to execute. Here’s the full hands-on video: 👇
1:30

🧵 1/7 I gave a Flash model a written spec for a 3D physics game. It broke the task into steps, wrote the code, used the tools, ran the project, and eventually produced a billiards game I could actually play. Ant Ling Ling-3.0-flash has 124B total parameters, but activates just 5.1B per token. That low activation count didn’t limit it to a simple one-shot demo. It kept a multi-stage 3D build moving from spec to playable result. This test made one thing clear: stop using your biggest model for every step of an agent workflow. The project may be complex. Most of the execution steps inside it don’t need a heavyweight planner to rethink everything from scratch. Ling-3.0-flash is built to execute. Here’s the full hands-on video: 👇

Lonely

119,450 views • 13 days ago

BREAKING NEWS 🚨 Google just released Video upload option in Gemini App. Already got access to it in my PRO plan. Check out this cool demo where I analyze a video in seconds.
0:47

BREAKING NEWS 🚨 Google just released Video upload option in Gemini App. Already got access to it in my PRO plan. Check out this cool demo where I analyze a video in seconds.

Prashant

18,947 views • 1 year ago

Holy moly! The new video feature in @NotebookLM is out of this world! This is AGI for teaching and education! Google Gemini really cooked with this! Right now you only get access with an Ultra account. Check out this video on my beloved T cells! It explains complex issues so well!
8:19

Holy moly! The new video feature in @NotebookLM is out of this world! This is AGI for teaching and education! Google Gemini really cooked with this! Right now you only get access with an Ultra account. Check out this video on my beloved T cells! It explains complex issues so well!

Derya Unutmaz, MD

146,816 views • 4 months ago

Humanoid robots are cool. But most of them aren’t ready for real-world work. X-Humanoid just launched Embodied Tien Kung 3.0 - a full-size humanoid robot built to actually operate in industrial and commercial environments. What makes it different? • Strong, high-torque joints for heavy tasks • Can move dynamically and stay stable • Handles complex motions (even over obstacles) • Millimeter-level precision for detailed work It runs on the Wise KaiWu AI platform, which allows the robot to see, think, and act in a continuous loop with less human control and support for multi-robot collaboration. It also supports ROS2, MQTT, and TCP/IP, and offers more open hardware + software interfaces to make development easier for teams. The goal is simple: make humanoid robots practical, easier to build on, and ready for real-world use. Visit: Check out the official launch video below to see what it can do! 👇 #Robotics #HumanoidRobot #AI
2:18

Humanoid robots are cool. But most of them aren’t ready for real-world work. X-Humanoid just launched Embodied Tien Kung 3.0 - a full-size humanoid robot built to actually operate in industrial and commercial environments. What makes it different? • Strong, high-torque joints for heavy tasks • Can move dynamically and stay stable • Handles complex motions (even over obstacles) • Millimeter-level precision for detailed work It runs on the Wise KaiWu AI platform, which allows the robot to see, think, and act in a continuous loop with less human control and support for multi-robot collaboration. It also supports ROS2, MQTT, and TCP/IP, and offers more open hardware + software interfaces to make development easier for teams. The goal is simple: make humanoid robots practical, easier to build on, and ready for real-world use. Visit: Check out the official launch video below to see what it can do! 👇 #Robotics #HumanoidRobot #AI

Chidanand Tripathi

69,604 views • 5 months ago

The new hotness in humanoid robotics! German robotics company NEURA has released a video of their humanoid robot, 4NE-1, performing tasks. The robot's concept was unveiled in 2022. NEURA is among the first to join the early access NVIDIA Humanoid Robot Developer Program.
0:32

The new hotness in humanoid robotics! German robotics company NEURA has released a video of their humanoid robot, 4NE-1, performing tasks. The robot's concept was unveiled in 2022. NEURA is among the first to join the early access NVIDIA Humanoid Robot Developer Program.

The Humanoid Hub

100,874 views • 2 years ago

This looks like another step toward removing the data bottleneck in robotics. The bottleneck exists because recording demonstrations is easy, but labeling every grasp, hold, move, and release is slow and expensive. Perceptron Egocentric dropped a new robotics video annotation system. You give it raw robot-camera or first-person/egocentric video. It gives back structured labels that look more like training data: where each small manipulation starts and ends, what the subtask is, what each hand is doing, where the hands are frame by frame, and which hand is left or right. So, Input: raw video of a person or robot doing a task. Output: a machine-readable breakdown of the physical actions inside that video. Basically, it is a machine that converts robot video into policy-training supervision. That means timestamps, subtask boundaries, per-hand actions, and left-right hand grounding. So this will turn large piles of robot or egocentric video into supervision for training robot policies.
0:58

This looks like another step toward removing the data bottleneck in robotics. The bottleneck exists because recording demonstrations is easy, but labeling every grasp, hold, move, and release is slow and expensive. Perceptron Egocentric dropped a new robotics video annotation system. You give it raw robot-camera or first-person/egocentric video. It gives back structured labels that look more like training data: where each small manipulation starts and ends, what the subtask is, what each hand is doing, where the hands are frame by frame, and which hand is left or right. So, Input: raw video of a person or robot doing a task. Output: a machine-readable breakdown of the physical actions inside that video. Basically, it is a machine that converts robot video into policy-training supervision. That means timestamps, subtask boundaries, per-hand actions, and left-right hand grounding. So this will turn large piles of robot or egocentric video into supervision for training robot policies.

Rohan Paul

18,903 views • 1 month ago

Watch as Sharpa North assembles a computer autonomously! From precise positioning to screwing and wire clipping, North executes complex, contact-rich tasks powered by tactile feedback. Dexterous hands + tactile AI are what make real-world task execution possible. This is the shift from simple automation to real-world robotics. #SharpaRobotics
1:56

Watch as Sharpa North assembles a computer autonomously! From precise positioning to screwing and wire clipping, North executes complex, contact-rich tasks powered by tactile feedback. Dexterous hands + tactile AI are what make real-world task execution possible. This is the shift from simple automation to real-world robotics. #SharpaRobotics

Sharpa

28,164 views • 4 months ago

This is a single uncut video, showing a robot learning several tasks instantly, after just one demonstration each ... This is possible because we've now been able to achieve in-context learning for everyday robotics tasks, and I'm very excited to announce our latest paper: 🎆 Instant Policy: In-Context Imitation Learning via Graph Diffusion 🎆 (1/6) 🧵👇
0:50

This is a single uncut video, showing a robot learning several tasks instantly, after just one demonstration each ... This is possible because we've now been able to achieve in-context learning for everyday robotics tasks, and I'm very excited to announce our latest paper: 🎆 Instant Policy: In-Context Imitation Learning via Graph Diffusion 🎆 (1/6) 🧵👇

Edward Johns

74,683 views • 1 year ago

You need to be using Gemini 3.0, trust me 1) Unmatched Intelligence - beats both ChatGPT and Claude on every benchmark 2) One-shots everything you give it 3) Massive Context - Process 1M tokens instantly 4) Speed - Near-zero latency for real-time workflows 5) Agentic Power - It executes tasks, not just text In this 19 min video, you will learn everything about mastering Gemini 3.0 (even as a beginner)
19:03

You need to be using Gemini 3.0, trust me 1) Unmatched Intelligence - beats both ChatGPT and Claude on every benchmark 2) One-shots everything you give it 3) Massive Context - Process 1M tokens instantly 4) Speed - Near-zero latency for real-time workflows 5) Agentic Power - It executes tasks, not just text In this 19 min video, you will learn everything about mastering Gemini 3.0 (even as a beginner)

David Ondrej

27,892 views • 8 months ago