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。

Today Google DeepMind released a state of the art computer use model, in partnership with Browserbase. Computer use is hard to evaluate. You need reliable browser infrastructure and realistic tasks. Below, we cover how we ran these benchmarks and how you can try yourself!

Browserbase

21,529 subscribers

155,120 次观看 • 11 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

29 条评论

Browserbase 的头像

Browserbase11 个月前

Read about the evaluations:

Logan Kilpatrick 的头像

Logan Kilpatrick11 个月前

@GoogleDeepMind Awesome to partner with you all and excited to get this new model into the hands of more builders!!!

Paul Klein IV 的头像

Paul Klein IV11 个月前

@GoogleDeepMind need to buy some Google stock after this

Kyle Jeong 的头像

Kyle Jeong11 个月前

@GoogleDeepMind Play with the models here

Kyle Jeong 的头像

Kyle Jeong11 个月前

@GoogleDeepMind if you're not using Browserbase to train your computer use models, you're not doing it right

Stella Wang 的头像

Stella Wang11 个月前

@GoogleDeepMind 🤺

derek 的头像

derek11 个月前

@GoogleDeepMind google us 😎

Stella Wang 的头像

Stella Wang11 个月前

@GoogleDeepMind Vibe evaluating🤔🤔

Linda Chen 的头像

Linda Chen11 个月前

@GoogleDeepMind Sickest crossover of all time

Peyton Casper 的头像

Peyton Casper11 个月前

@GoogleDeepMind can't believe we built @mercor in 24hrs

Erin Goode 的头像

Erin Goode11 个月前

@GoogleDeepMind let's goooooo

sameel arif 的头像

sameel arif11 个月前

@GoogleDeepMind 🅱️ig moves

harsehaj ⋆˙⟡ 的头像

harsehaj ⋆˙⟡11 个月前

@GoogleDeepMind reliable browser infra !!!

alana goyal 的头像

alana goyal11 个月前

@GoogleDeepMind this is big

Philipp Schmid 的头像

Philipp Schmid11 个月前

@GoogleDeepMind Awesome work! 🚀

Filip Michalsky 的头像

Filip Michalsky11 个月前

@GoogleDeepMind This is HUGE

Sheing Ng 的头像

Sheing Ng11 个月前

@GoogleDeepMind Ping me again if it's able to solve a wordle challenge.

Stella Wang 的头像

Stella Wang11 个月前

@GoogleDeepMind THIS IS INSANE

⟁ndrew V 的头像

⟁ndrew V11 个月前

@GoogleDeepMind That’s awesome. What an epic feeling that must be to have a partnership like this announcement for y’all. Also looking forward to trying it and building something great with the potential!

Junaid 的头像

Junaid11 个月前

@GoogleDeepMind Amazing stuff going on at 🅱️

Aman DevX 的头像

Aman DevX11 个月前

@GoogleDeepMind That sounds pretty cool, Browserbase!

tim 的头像

tim11 个月前

@GoogleDeepMind is it just me, or is it a bit confusing on what to do after agent completes the task? I was expecting the result, but couldn't find that anywhere on the UI

parth 的头像

parth11 个月前

@GoogleDeepMind 🙌

Sughu 的头像

Sughu11 个月前

@GoogleDeepMind Please create a pay per use API tier

Nikhil Kulkarni 的头像

Nikhil Kulkarni11 个月前

@GoogleDeepMind lets goooo!

Turner Novak 🍌🧢 的头像

Turner Novak 🍌🧢11 个月前

@GoogleDeepMind Wow

lazy_procrastinator 的头像

lazy_procrastinator11 个月前

@GoogleDeepMind Genuinely curious, why does one need browserbase for this? Doesn’t it work headless?

Collin Shill 的头像

Collin Shill11 个月前

@GoogleDeepMind LFG

Ethan 的头像

Ethan11 个月前

@GoogleDeepMind Nice

相关视频

Computer use is finally reliable. It just needed a real harness. Now there is one. Starting today, anyone can ship a SOTA browser agent, batteries included. Introducing Browserbase Agents: one prompt and one API call is all you need to automate the whole web.

Computer use is finally reliable. It just needed a real harness. Now there is one. Starting today, anyone can ship a SOTA browser agent, batteries included. Introducing Browserbase Agents: one prompt and one API call is all you need to automate the whole web.

Browserbase

96,484 次观看 • 3 个月前

We've partnered with OpenAI as one of their early research preview partners for the new computer use model. With this model, AI can control a browser better than ever before 👀 Of course, we had to launch a free hosted version + open source implementation. Check it out!

We've partnered with OpenAI as one of their early research preview partners for the new computer use model. With this model, AI can control a browser better than ever before 👀 Of course, we had to launch a free hosted version + open source implementation. Check it out!

Browserbase

40,232 次观看 • 1 年前

A look at computer use and the built-in browser in the new ChatGPT app 👀 dominik kundel walks through how ChatGPT can work with apps on your computer and browse the web to research, navigate websites, and complete tasks with you.

A look at computer use and the built-in browser in the new ChatGPT app 👀 dominik kundel walks through how ChatGPT can work with apps on your computer and browse the web to research, navigate websites, and complete tasks with you.

ChatGPT

128,615 次观看 • 2 个月前

You've probably heard of us before, but what exactly is Browserbase? Here's everything you need to know about how we can help you automate the web, in 90 seconds.

You've probably heard of us before, but what exactly is Browserbase? Here's everything you need to know about how we can help you automate the web, in 90 seconds.

Browserbase

39,801 次观看 • 9 个月前

Forgot to make dinner plans? No worries, you can now use the Browserbase MCP Server with Raycast to book a fancy place in no time. The everything MCP. Try it out now! Config below.

Forgot to make dinner plans? No worries, you can now use the Browserbase MCP Server with Raycast to book a fancy place in no time. The everything MCP. Try it out now! Config below.

Browserbase

34,963 次观看 • 1 年前

HoloTab is officially live. 🚀 14 days ago, we released Holo3. It proved you can hit state-of-the-art computer-use performance, outperforming the biggest models in the world at 1/10th of the cost. Now, we are moving that power out of the research labs and into your browser tabs. Don't trust the benchmarks. Trust your own experience! #AI #AgenticAI #HoloTab #ComputerUse #Innovation #HCompany

HoloTab is officially live. 🚀 14 days ago, we released Holo3. It proved you can hit state-of-the-art computer-use performance, outperforming the biggest models in the world at 1/10th of the cost. Now, we are moving that power out of the research labs and into your browser tabs. Don't trust the benchmarks. Trust your own experience! #AI #AgenticAI #HoloTab #ComputerUse #Innovation #HCompany

H

25,183 次观看 • 5 个月前

Your agents should use a remote browser built specifically for them, We revamped our Browser Skill to use our custom Stagehand CLI, It's faster than MCP tools and each run uses a new browser context, avoiding prompt injection. npx skills add browserbase/skills —skill browser

Your agents should use a remote browser built specifically for them, We revamped our Browser Skill to use our custom Stagehand CLI, It's faster than MCP tools and each run uses a new browser context, avoiding prompt injection. npx skills add browserbase/skills —skill browser

Browserbase

41,144 次观看 • 6 个月前

Google just unveiled a new project that shows how AI Agents are coming for full browser and computer use. This opens up AI to any of the laborious knowledge tasks that we do today. Most importantly, it lets us automate work that we never had even contemplated before.

Google just unveiled a new project that shows how AI Agents are coming for full browser and computer use. This opens up AI to any of the laborious knowledge tasks that we do today. Most importantly, it lets us automate work that we never had even contemplated before.

Aaron Levie

145,442 次观看 • 1 年前

OpenClaw is all the rage and skills are a huge reason why. We've revamped ours to make it easier for Claude to go from local to prod. Check out how you can use Claude, Skills, and GitHub Actions to automatically self-heal browser automation scripts.

OpenClaw is all the rage and skills are a huge reason why. We've revamped ours to make it easier for Claude to go from local to prod. Check out how you can use Claude, Skills, and GitHub Actions to automatically self-heal browser automation scripts.

Browserbase

27,910 次观看 • 7 个月前

Today we're excited to introduce Vy, our AI that sees and acts on your computer. At Vercept, our mission is to reinvent how humans use computers–enabling you to accomplish orders of magnitude more than what you can do today. Vy is a first glimpse at AI that sees and uses your computer just like you do. It runs natively on your computer, with access to all your applications and accounts that you’re already signed into. It is available for download starting today! As part of the release, we’ve built VyUI, an AI model bridging the gap between language and your screen. It sets a new state-of-the-art on UI grounding benchmarks like ScreenSpot v1, ScreenSpot v2, and GroundUI Web–beating leading models from OpenAI, Google, and Anthropic. Check out what Vy can do in the thread below: 🧵👇

Today we're excited to introduce Vy, our AI that sees and acts on your computer. At Vercept, our mission is to reinvent how humans use computers–enabling you to accomplish orders of magnitude more than what you can do today. Vy is a first glimpse at AI that sees and uses your computer just like you do. It runs natively on your computer, with access to all your applications and accounts that you’re already signed into. It is available for download starting today! As part of the release, we’ve built VyUI, an AI model bridging the gap between language and your screen. It sets a new state-of-the-art on UI grounding benchmarks like ScreenSpot v1, ScreenSpot v2, and GroundUI Web–beating leading models from OpenAI, Google, and Anthropic. Check out what Vy can do in the thread below: 🧵👇

Vercept

123,859 次观看 • 1 年前

Just found an open-source tool that lets you see how well an AI can actually use a computer. it’s called OSWorld 2.0. you give the AI a real task on a computer, then watch to see if it can figure everything out and finish it. it gives you: → real computer tasks, not simple questions → tasks that can take many steps to finish → the same computer setup for every AI → websites made for testing → all the files and tasks for free this is also the environment used to test computer-use systems from OpenAI and Anthropic. so if you’re building AI agents, researching computer use, or just want to see what these models can actually do, you can run the same tests yourself instead of relying only on benchmark scores. the labs publish the scores. this repo is where the actual tests come from. bookmark this.

Just found an open-source tool that lets you see how well an AI can actually use a computer. it’s called OSWorld 2.0. you give the AI a real task on a computer, then watch to see if it can figure everything out and finish it. it gives you: → real computer tasks, not simple questions → tasks that can take many steps to finish → the same computer setup for every AI → websites made for testing → all the files and tasks for free this is also the environment used to test computer-use systems from OpenAI and Anthropic. so if you’re building AI agents, researching computer use, or just want to see what these models can actually do, you can run the same tests yourself instead of relying only on benchmark scores. the labs publish the scores. this repo is where the actual tests come from. bookmark this.

MIKE

14,387 次观看 • 17 天前

We're excited to announce our partnership with Prime Intellect to allow anyone to train browser agents. General-purpose models aren't optimized for your browser workflows, BrowserEnv lets you train one that is. Checkout and train your own custom model in a few hours.

We're excited to announce our partnership with Prime Intellect to allow anyone to train browser agents. General-purpose models aren't optimized for your browser workflows, BrowserEnv lets you train one that is. Checkout and train your own custom model in a few hours.

Browserbase

191,891 次观看 • 6 个月前

We just built AWS Lambda with a browser built-in. Introducing Functions, a new compute paradigm that co-locates your code and browsers to build faster browser agents. Functions lets your team focus on the product, while we handle the infrastructure.

We just built AWS Lambda with a browser built-in. Introducing Functions, a new compute paradigm that co-locates your code and browsers to build faster browser agents. Functions lets your team focus on the product, while we handle the infrastructure.

Browserbase

63,284 次观看 • 7 个月前

After coding is solved, the next frontier is computer use. Today, we are launching Use Computer, the infra for evaluating and training models to use all kinds of computers 👇

After coding is solved, the next frontier is computer use. Today, we are launching Use Computer, the infra for evaluating and training models to use all kinds of computers 👇

Marco Mascorro

44,641 次观看 • 3 个月前

Claude Fable 5 is now available in Computer as an orchestrator model. This is Anthropic's state-of-the-art model for long, complex tasks. Available only to Pro and Max subscribers in Computer.

Claude Fable 5 is now available in Computer as an orchestrator model. This is Anthropic's state-of-the-art model for long, complex tasks. Available only to Pro and Max subscribers in Computer.

Perplexity

139,671 次观看 • 3 个月前

For our European friends 🎉 we are rolling out this week Codex for Chrome, Computer Use, Memories, and Chronicle in EU, EEA and UK. Here's a recap on how you can use Codex with Chrome and some more tips below 🧵

For our European friends 🎉 we are rolling out this week Codex for Chrome, Computer Use, Memories, and Chronicle in EU, EEA and UK. Here's a recap on how you can use Codex with Chrome and some more tips below 🧵

dominik kundel

25,555 次观看 • 3 个月前

Papers With Code SOTA is back 🚀 We used Gemini 3 to process millions of charts and tables across arXiv + the web to surface state-of-the-art AI research in reasoning, computer-use, OCR, and more Track the best-performing methods and see which benchmarks are getting adopted

Papers With Code SOTA is back 🚀 We used Gemini 3 to process millions of charts and tables across arXiv + the web to surface state-of-the-art AI research in reasoning, computer-use, OCR, and more Track the best-performing methods and see which benchmarks are getting adopted

alphaXiv

17,856 次观看 • 10 个月前

🎮 Computer Use Agent Arena is LIVE! 🚀 🔥 Easiest way to test computer-use agents in the wild without any setup 🌟 Compare top VLMs: OpenAI Operator, Claude 3.7, Gemini 2.5 Pro, Qwen 2.5 vl and more 🕹️ Test agents on 100+ real apps & webs with one-click config 🔒 Safe & free access on cloud-hosted machines Page: Leaderboard (tentative): Blog: Data & Code (coming soon): ⭐️Why Computer Agent Arena? 1️⃣Beyond Static Benchmarks: We use computers to perform enormous tasks and workflows every day, and AI agents have the potential to automate these tasks. However, existing benchmarks are very limited (e.g., only 369 tasks in OSWorld and 812 tasks in WebArena). To better measure their capabilities, we introduce Computer Agent Arena for users to easily compare & test AI agents on all kinds of crowdsourced real-world computer use tasks. 2️⃣Cloud Testing, Simplified: As agents like OpenAI’s Operator and Claude 3.7 sonnet release, users face configuration challenges and privacy hurdles to deploy on their own computers. Our platform integrates these agents with cloud-hosted machines, providing users with quick and secure access. 3️⃣Unified Embodied Digital Environment: Unlike Chatbot Arena, we provide users with a real embodied environment—computers—where all agents are grounded in real computer tasks and environments. Led by @XLANG_Lab [1/🧵]

🎮 Computer Use Agent Arena is LIVE! 🚀 🔥 Easiest way to test computer-use agents in the wild without any setup 🌟 Compare top VLMs: OpenAI Operator, Claude 3.7, Gemini 2.5 Pro, Qwen 2.5 vl and more 🕹️ Test agents on 100+ real apps & webs with one-click config 🔒 Safe & free access on cloud-hosted machines Page: Leaderboard (tentative): Blog: Data & Code (coming soon): ⭐️Why Computer Agent Arena? 1️⃣Beyond Static Benchmarks: We use computers to perform enormous tasks and workflows every day, and AI agents have the potential to automate these tasks. However, existing benchmarks are very limited (e.g., only 369 tasks in OSWorld and 812 tasks in WebArena). To better measure their capabilities, we introduce Computer Agent Arena for users to easily compare & test AI agents on all kinds of crowdsourced real-world computer use tasks. 2️⃣Cloud Testing, Simplified: As agents like OpenAI’s Operator and Claude 3.7 sonnet release, users face configuration challenges and privacy hurdles to deploy on their own computers. Our platform integrates these agents with cloud-hosted machines, providing users with quick and secure access. 3️⃣Unified Embodied Digital Environment: Unlike Chatbot Arena, we provide users with a real embodied environment—computers—where all agents are grounded in real computer tasks and environments. Led by @XLANG_Lab [1/🧵]

Bowen Wang

93,273 次观看 • 1 年前