Hacker News new | ask | show | jobs
by mdrzn 851 days ago
"Use any LLM of your choice like OpenAI, Claude or Gemeni Pro" will local LLMs like LLaMa/LLaVa be added in the future?

edit: nevermind, I thought it was fully open source, instead you will still have to pay for "videodb" on top of your AI tokens.
1 comments
ashu_trv 851 days ago
Yeah, VideoDB is the next-gen infrastructure for videos and actually less costly than current video infrastructure.

But you can use any LLM for analysing the transcript.

link
mdrzn 851 days ago
So it's still just analysing the transcript, not using GPT4-V or OCR in any way?

Can you confirm if I could skip using VideoDB by using Whisper to transcribe the video, and then use that transcript with LLaMa to extract the important parts?

link
ashu_trv 851 days ago
It analyse the transcript, but there is no way to get back the video clip without building your own video infra. We at Videodb are solving the exact problem.
link