Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Deep Research wasn’t fast enough, so we built Wide Research: multi-agent web research powered by Scrapybara Act SDK Watch CUA navigate socials to find YC W25 contact infos - even when limited to parallel batches of 4 it researched 1 company/min! gib rate limit pls edwin

15,854 Aufrufe • vor 1 Jahr •via X (Twitter)

8 Kommentare

Profilbild von justin
justinvor 1 Jahr

Build with CUA:

Profilbild von AssemblyAI
AssemblyAIvor 1 Jahr

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 👇

Profilbild von justin
justinvor 1 Jahr

To clarify our infrastructure supports researching every company at the same time, only limiting the batch size rn because of LLM rate limits!

Profilbild von Kyle Jeong
Kyle Jeongvor 1 Jahr

@edwinarbus when are we getting obese research (100 concurrent vms)

Profilbild von justin
justinvor 1 Jahr

@edwinarbus it’s coming

Profilbild von Houdini
Houdinivor 1 Jahr

@edwinarbus Where can I try this out. Ready to pay

Profilbild von Kushagra
Kushagravor 1 Jahr

@edwinarbus @utkarshcs18

Profilbild von Oleg Zaremba
Oleg Zarembavor 1 Jahr

@edwinarbus Why did you split it into 4 only after getting the list? Also, how many "steps-per-prompt" were you using? Did you optimize for that?

Ähnliche Videos