Видео

Thank you so much Jason this amazing, but i keep getting this issue The agent 'user_proxy' has an empty system_message, and may not work well with GroupChat

But the don’t count childhood adaptation to world, and studying and those patterns , that takes several years

The title is confusing, wouldn't have clicked if i wasn't familiar with your content

that intro HITS so harddd 😭👌🏼

Nice! Is there a reason you don’t try using langgraph for multi agent frameworks?

Why not do a brain scan on someone solving this problem and then make a short hand representation of that reaction in code? Apologies if this is am unhinged idea, I've just awaken.

the latest one is 39% now! have anyone tried finetuning?

Gonna give it a try, I compared Claude 3.5 VS gpt4o, claude 3.5 is clearly better; though still struggle

This is my new favorite channel. The topics are pretty dense and dry - but you make them super easy and fun to learn. Thank you!

am i trippin or did i just see someone walk past behind you? 4:35

I'm getting an error at line 17 db=FAISS.from_docu... 404 error. not sure how to resolve that.

> get_updates()

I have been thinking of an idea where websites have a ***/chat endpoint where all the website data that is relevant has been converted to embeddings for extern system LLM consumption. The website owner could either host this service themselves, or a 3d party. This could be a way for website owner to increase traffic since they went to trouble to already "LLMize" their site. Or end user could just"chat" with website and skip past UI design

My screen was like: 🌚🌝🌚🌝🌚😀👀😵 Why not a dark theme on hugging face too?

Really the best RAG video on youtube so far

Hi Jason, I m a Mac Silicon user, followed all the steps till adding the 'Prepare Image for InsightFace' Node, cant find it in the list, i can only see 'IPAdapter InsightFace Loader', please help, thank you.

I’ve got a question and i need a real answer. Am I a nerd for being interested in this because I try to talk to people about this and no one seems to be even remotely interested.

I am getting the error "ValueError: Unable to create tensor, you should probably activate padding with 'padding=True' to have batched tensors with the same length." when using the image to text model

can i run the code in free collab

what gpu is needed?

How to protect a company's information with this technology?

Shitty tutorial.

what is your opinion on using hypothetical answer before search? e.g. send query to llm to get general answer, which is used to enrich the query to get more context, then send enriched query to vector store to do a search

Am I the only one who thought that was a banger

great stuff, as new to hearing this, very interesting, can this be built by a novice . . .

Excellent idea and implementation!

😊😊

Man! You're much better than AI! 😀

AI drum and bass when

Your personality shines through in every video you make!

you have your example on GitHub ? thanks

16:38 ..... "badeed"

you won a new subscriber with this superb video! thanks a lot for sharing all the know how

Nice work Jason!

genai.upload_file and get_file is not working for some reason

lost in the middle seems not a problem on the large ctx of claude and gemini

As always, quality research

Such a great video. You're great Jason. One of my best subs on RUclips

Suno is garbage, too many artifacts.

udio is actually mind blowing

Reupload?

Reupload?

api is probably using session tokens of free accounts they created. I wouldn't build a serious app with it.

Suno has an API? 🤔

reupload?

Whoa, the audio/video geneeration came a long way past few month! cant imagining what it will be in like 5 month...

Suno sounds way too robotic it's like they are singing into a tin can, Udio sounds more real, have you tried it?

I'm thinking about also using elevan labs to build an app for audio effect generation based on video clips; this demo is a great reference!

Didn't know suno has API, this is great; gonna try it out before this gonna banned!

First