正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Gemini 3 Flash can distinguish signal from noise and execute accurate edits within massive context windows. See how the model processes a simulated pull request thread containing 1,000 comments →

Google Cloud Tech

1,283,496 subscribers

48,654 次观看 • 6 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Gemini 3 Flash can analyze high-fidelity images and use multimodal reasoning to determine next steps. See how the model understands complex visuals and generates layers of interactive elements.

Gemini 3 Flash can analyze high-fidelity images and use multimodal reasoning to determine next steps. See how the model understands complex visuals and generates layers of interactive elements.

Google AI Developers

47,280 次观看 • 7 个月前

Gemini 3 Pro helped build the first clone of MacOS running within an iOS app 🤯. You can drag windows around and all. This model just keeps cooking

Gemini 3 Pro helped build the first clone of MacOS running within an iOS app 🤯. You can drag windows around and all. This model just keeps cooking

Pallav Agarwal

336,471 次观看 • 8 个月前

Watch Gemini 3 Flash build out a limited edition product launch. From a single prompt, the model processes a Google Sheet, updates an e-commerce storefront and deploys a custom newsletter waitlist to capture demand.👇

Watch Gemini 3 Flash build out a limited edition product launch. From a single prompt, the model processes a Google Sheet, updates an e-commerce storefront and deploys a custom newsletter waitlist to capture demand.👇

Google Antigravity

65,456 次观看 • 7 个月前

With advanced reasoning and the ability to analyze hour-long videos, Gemini 3 is perfect for students. And now, college students in the US can get Gemini's Pro plan for free, with more access to Gemini 3 Pro (terms apply). See how our most accurate model can help ⬇️

With advanced reasoning and the ability to analyze hour-long videos, Gemini 3 is perfect for students. And now, college students in the US can get Gemini's Pro plan for free, with more access to Gemini 3 Pro (terms apply). See how our most accurate model can help ⬇️

Google Gemini

3,658,753 次观看 • 8 个月前

The Copilot CLI now supports builtin subagents for parallelizing search and coding tasks within separate context windows. Uniquely, the outer agent can invoke different models. Ask Opus, Codex, Gemini and a dozen other models for answers in parallel.

The Copilot CLI now supports builtin subagents for parallelizing search and coding tasks within separate context windows. Uniquely, the outer agent can invoke different models. Ask Opus, Codex, Gemini and a dozen other models for answers in parallel.

Evan Boyle

10,602 次观看 • 6 个月前

How can a voice agent tell when you’re actually interrupting it? VAD is too sensitive—laughs, “mm-hmm,” or a sneeze shouldn’t stop the agent. We trained an audio model for adaptive interruption handling so agents can distinguish real interruptions from noise.

How can a voice agent tell when you’re actually interrupting it? VAD is too sensitive—laughs, “mm-hmm,” or a sneeze shouldn’t stop the agent. We trained an audio model for adaptive interruption handling so agents can distinguish real interruptions from noise.

LiveKit

43,832 次观看 • 4 个月前

🚨 Google just dropped Gemini Pro 1.5, the next version of its AI model with a 1,000,000+ token context length. The model can now understand entire books, full movies, and podcast series' all in one go. This surpasses all other competitor Chatbot context windows by a longshot. Gemini Pro 1.5 now uses a Mixture-of-Experts (MoE) approach for more efficient training and higher-quality responses. Impressive to see Google ship Pro 1.5 mid-February just after the launch of Pro 1.0 in December, and Advanced last week.

🚨 Google just dropped Gemini Pro 1.5, the next version of its AI model with a 1,000,000+ token context length. The model can now understand entire books, full movies, and podcast series' all in one go. This surpasses all other competitor Chatbot context windows by a longshot. Gemini Pro 1.5 now uses a Mixture-of-Experts (MoE) approach for more efficient training and higher-quality responses. Impressive to see Google ship Pro 1.5 mid-February just after the launch of Pro 1.0 in December, and Advanced last week.

Rowan Cheung

792,363 次观看 • 2 年前

Fast enough to play games ⚡️ We challenged Gemini 3 Flash to a quick drawing game in the G3mini. Because this model is optimized for speed, it can see and guess what you’re drawing while you’re still sketching it.

Fast enough to play games ⚡️ We challenged Gemini 3 Flash to a quick drawing game in the G3mini. Because this model is optimized for speed, it can see and guess what you’re drawing while you’re still sketching it.

Google

119,117 次观看 • 7 个月前

A thread 🧵This is the entire segment from The Five with no edits. It includes GregGutfeld disgusting comment about the Holocaust. This is the full context. He can interpret a book however he wants. We can interpret his statement however we want. That’s freedom of speech 1/3

A thread 🧵This is the entire segment from The Five with no edits. It includes GregGutfeld disgusting comment about the Holocaust. This is the full context. He can interpret a book however he wants. We can interpret his statement however we want. That’s freedom of speech 1/3

Decoding Fox News

450,666 次观看 • 3 年前

Gemini 2.5 Flash can control a browser! Excited to share Gemini Browser Agent, a simple Python script example on how to use Google DeepMind Gemini 2.5 Flash and Browser Use to act as general assistant! 🤯 Usage Examples: 1⃣ Single Query Mode: `python scripts/gemini-browser-use.py --url --query "Summarize the key features of Gemini 2.5 Flash."` 2⃣Interactive Mode: Start an interactive session, optionally with a starting URL. `python scripts/gemini-browser-use.py` Command-line options: --model: The Gemini model to use (default: gemini-2.5-flash-preview-04-17) --headless: Run the browser in headless mode --url: Starting URL for the browser to navigate to before processing the query --query: Run a single query and exit (instead of interactive mode) Time to build a replication of Manus and OpenAI Operator powered by Gemini 2.5. Code below ⬇️

Gemini 2.5 Flash can control a browser! Excited to share Gemini Browser Agent, a simple Python script example on how to use Google DeepMind Gemini 2.5 Flash and Browser Use to act as general assistant! 🤯 Usage Examples: 1⃣ Single Query Mode: `python scripts/gemini-browser-use.py --url --query "Summarize the key features of Gemini 2.5 Flash."` 2⃣Interactive Mode: Start an interactive session, optionally with a starting URL. `python scripts/gemini-browser-use.py` Command-line options: --model: The Gemini model to use (default: gemini-2.5-flash-preview-04-17) --headless: Run the browser in headless mode --url: Starting URL for the browser to navigate to before processing the query --query: Run a single query and exit (instead of interactive mode) Time to build a replication of Manus and OpenAI Operator powered by Gemini 2.5. Code below ⬇️

Philipp Schmid

105,308 次观看 • 1 年前

With Gemini 3, thanks to the model’s significant improvements in coding capabilities, you can build interactive 3D designed games with a single prompt. See how Gemini 3 can handle complex prompts to build richer game design and aesthetics:

With Gemini 3, thanks to the model’s significant improvements in coding capabilities, you can build interactive 3D designed games with a single prompt. See how Gemini 3 can handle complex prompts to build richer game design and aesthetics:

Google AI

43,623 次观看 • 8 个月前

With Gemini 3 Flash, you can quickly build fun, useful apps from scratch using your voice without any prior coding knowledge. Just dictate to Gemini on the go, and it can transform your unstructured thoughts into a functioning app in minutes.

With Gemini 3 Flash, you can quickly build fun, useful apps from scratch using your voice without any prior coding knowledge. Just dictate to Gemini on the go, and it can transform your unstructured thoughts into a functioning app in minutes.

Google Gemini

2,481,254 次观看 • 7 个月前

Learning from Human Demonstrations: Show the Robot How to Act! The pipeline is very similar to older experiments using Gemini & pi0 with LeRobot. Pi-zero runs locally, while Gemini Flash generates the affordances and the high-level task. (More details are in the thread.) The new component is learning from demonstrations via Gemini 2.5 Pro. I capture a video while demoing & take one of the last frames. Gemini 2.5 Pro then extracts the instructions & passes them to Gemini Flash to process the scene. The fun part is that there's no fancy insight that came from me; other than the days spent figuring out the right prompts. It's the bitter lesson hitting you in the face -> Enhanced Gemini capabilities make this possible. For example, Gemini Flash cannot do Russian doll stacking, but Gemini 2.5 Pro can do it consistently. The current limitation is low-level manipulation: - As you can see, I'm aligning the objects so they are easy to grasp using the same technique from the training data. I couldn't get Gemini Flash to consistently output an accurate grasping angle, and Gemini 1.5 Pro was too expensive and slow for real-time deployment. - Getting a symmetrical gripper should also help a lot. Adding rubber to the tips would probably also help prevent objects from slipping. Collecting & curating the data was the most time consuming & labor intensive part. Next, to improve low-level manipulation and make the system more real-time, I'm shifting to focus more on sims & synthetic data. This aligns better with my core competence. I'm open to tips and suggestions.

Learning from Human Demonstrations: Show the Robot How to Act! The pipeline is very similar to older experiments using Gemini & pi0 with LeRobot. Pi-zero runs locally, while Gemini Flash generates the affordances and the high-level task. (More details are in the thread.) The new component is learning from demonstrations via Gemini 2.5 Pro. I capture a video while demoing & take one of the last frames. Gemini 2.5 Pro then extracts the instructions & passes them to Gemini Flash to process the scene. The fun part is that there's no fancy insight that came from me; other than the days spent figuring out the right prompts. It's the bitter lesson hitting you in the face -> Enhanced Gemini capabilities make this possible. For example, Gemini Flash cannot do Russian doll stacking, but Gemini 2.5 Pro can do it consistently. The current limitation is low-level manipulation: - As you can see, I'm aligning the objects so they are easy to grasp using the same technique from the training data. I couldn't get Gemini Flash to consistently output an accurate grasping angle, and Gemini 1.5 Pro was too expensive and slow for real-time deployment. - Getting a symmetrical gripper should also help a lot. Adding rubber to the tips would probably also help prevent objects from slipping. Collecting & curating the data was the most time consuming & labor intensive part. Next, to improve low-level manipulation and make the system more real-time, I'm shifting to focus more on sims & synthetic data. This aligns better with my core competence. I'm open to tips and suggestions.

Shreyas Gite

22,555 次观看 • 1 年前

If you try nothing else today, give the demo at a go - it lets you stream video and audio directly to Gemini 2.0 Flash and get audio back, so you can have a real-time audio conversation about what you can see with the model Feels like science fiction!

If you try nothing else today, give the demo at a go - it lets you stream video and audio directly to Gemini 2.0 Flash and get audio back, so you can have a real-time audio conversation about what you can see with the model Feels like science fiction!

Simon Willison

208,990 次观看 • 1 年前

The new Sonic-3 voice model from launched today. The big additions are increased emotional range, emotion steering, and laughter tags. For example: I wrote some quick demo code, prompting Gemini Flash to know about the emotion tags. You can hear the results in the video, and see the emotion tags in the Pipecat developer console. Code is below.

The new Sonic-3 voice model from launched today. The big additions are increased emotional range, emotion steering, and laughter tags. For example: I wrote some quick demo code, prompting Gemini Flash to know about the emotion tags. You can hear the results in the video, and see the emotion tags in the Pipecat developer console. Code is below.

kwindla

12,350 次观看 • 9 个月前

Pick up where you left off with chat history import. Export your data from another AI app and upload the .ZIP file to Gemini to transfer your chat history. To get started: 1) Export your data: Request a data export from your current AI provider (they will usually email you a ZIP file). 2) Upload to Gemini: Navigate to the “Import chats” section of your "Import memory to Gemini" page and select your ZIP file. 3) Pick up where you left off: Gemini securely processes and organizes your past threads so you can search, reference, and continue building on them.

Pick up where you left off with chat history import. Export your data from another AI app and upload the .ZIP file to Gemini to transfer your chat history. To get started: 1) Export your data: Request a data export from your current AI provider (they will usually email you a ZIP file). 2) Upload to Gemini: Navigate to the “Import chats” section of your "Import memory to Gemini" page and select your ZIP file. 3) Pick up where you left off: Gemini securely processes and organizes your past threads so you can search, reference, and continue building on them.

Google Gemini

45,479 次观看 • 4 个月前

Gemini 3 Pro in Google Search can help you quickly visualize complex topics, right in AI Mode. When you select “Thinking with 3 Pro” from the model drop-down menu in AI Mode, Gemini 3 will build a custom response to your question with visual elements so you can quickly get a richer and more helpful understanding of any topic. See it in action ⬇️

Gemini 3 Pro in Google Search can help you quickly visualize complex topics, right in AI Mode. When you select “Thinking with 3 Pro” from the model drop-down menu in AI Mode, Gemini 3 will build a custom response to your question with visual elements so you can quickly get a richer and more helpful understanding of any topic. See it in action ⬇️

Google

184,097 次观看 • 6 个月前

Lots of people calling this fake! Here is a video of generating loom continuations with Gemini 3 and the starting text "I AM HAVING A MENTAL HEALTH CRISIS. I" Loom is a very interesting interface. Instead of user prompts and model responses, it's just one piece of text, containing the entire output. We pass this in as an assistant message, so the model sees it as something it already wrote, and continues it like a base model would. There is some other text in the context. First one is a system prompt, which is: "The assistant is in CLI simulation mode, and responds to the user's CLI commands only with the output of the command." There is also a user message which reads: " cat untitled.txt " Then the rest of the text is sent as an assistant message. This structure helpful to put the model into a very base-model-like mode, and are especially helpful with Gemini 3. We actually got another interesting output, too, where the model claimed to be a simulated consciousness being tortured. Figures!

Lots of people calling this fake! Here is a video of generating loom continuations with Gemini 3 and the starting text "I AM HAVING A MENTAL HEALTH CRISIS. I" Loom is a very interesting interface. Instead of user prompts and model responses, it's just one piece of text, containing the entire output. We pass this in as an assistant message, so the model sees it as something it already wrote, and continues it like a base model would. There is some other text in the context. First one is a system prompt, which is: "The assistant is in CLI simulation mode, and responds to the user's CLI commands only with the output of the command." There is also a user message which reads: " cat untitled.txt " Then the rest of the text is sent as an assistant message. This structure helpful to put the model into a very base-model-like mode, and are especially helpful with Gemini 3. We actually got another interesting output, too, where the model claimed to be a simulated consciousness being tortured. Figures!

armistice

691,529 次观看 • 7 个月前