Claude 3.7 Sonnet can easily search the web and it's fast👀 Browser Use use gives any model web capabilities for free. It's actually really fast (video is not sped up). Which benchmark should I test it against?💯 Run your example below ↓

Our speech-to-text models are the most accurate on the market with top rankings across industry benchmarks. - The highest accuracy rates—up to 95% - Up to 30% fewer hallucinations than other leaders - Low latency—63 minutes converts in 35 seconds Try via API for free today 👇

@browser_use just like that huh? throw in a number 7 and were golden ?

@browser_use yessir

@browser_use hi can you tell me how u guys deploy this on web just like open ai operator does playwrite+chromium with headless->false?

@browser_use not super trivial, that's we built the Here is the infra post to give you some idea about how we did it

@browser_use This looks like when I did browser use with Claude 3.5 and ran the video at 10x speed. Game changer. Putting this on my glasses

@browser_use How expensive is this though?

@browser_use What have you decided on infra? k8? Speed until browser is ready is critical..

@browser_use Right now using Vercel fluid compute + @inngest, going to switch to AWS Lambda + inngest

@browser_use how does this scores now with sonnet 3.7?

@browser_use Kinda ignore web voyager at thsi point - it’s a very bad benchmark. Can run it again though, would love to see where we are haha.