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!

Read about the evaluations:

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

@GoogleDeepMind need to buy some Google stock after this

@GoogleDeepMind Play with the models here

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

@GoogleDeepMind 🤺

@GoogleDeepMind google us 😎

@GoogleDeepMind Vibe evaluating🤔🤔

@GoogleDeepMind Sickest crossover of all time

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

@GoogleDeepMind let's goooooo

@GoogleDeepMind 🅱️ig moves

@GoogleDeepMind reliable browser infra !!!

@GoogleDeepMind this is big

@GoogleDeepMind Awesome work! 🚀

@GoogleDeepMind This is HUGE

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

@GoogleDeepMind THIS IS INSANE

@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!

@GoogleDeepMind Amazing stuff going on at 🅱️

@GoogleDeepMind That sounds pretty cool, Browserbase!

@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

@GoogleDeepMind 🙌

@GoogleDeepMind Please create a pay per use API tier

@GoogleDeepMind lets goooo!

@GoogleDeepMind Wow

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

@GoogleDeepMind LFG

@GoogleDeepMind Nice