Video

we want more videos on llama agentic system integration with the llama llm

Por fin alguien se dio cuenta, pero aun queda mas, gracias

this useless person just wasted my time by putting Langchain in video title

Clement Lake

Amazing tutorial, thank you Json

github link please

Nice! 🏆

Ok, you deserve a subscribe! You know what... usually I don't get fascinated by people teaching this stuff because I feel that they are redundant and just doing it for views. However, you were very smooth and went out from explaining it into doing a project in nearly 3 mins with explaining, so I have to thank you BRO! KEEP IT UP PLEASE, WE NEED PEOPLE LIKE YOU!!

Thanks for the video

Harris Jessica Young Richard Thompson Patricia

Can we do this with a shared google sheet across a team that maintains ongoing Q&As?

All the products and services and workflows you need to invest into and join... STUPID. All these ready made tools. I will spend the rest of my days finding out how to make it easy to make RAG local and private. I'm sick of everyone using python and overrelying on the ecosystem instead of the actual understanding. We are moving backwards. All these technical explanations are lowkey ignorant of how little this makes sense to people without context. Dont give me these services. Give me understanding and lowlevel tools. I'm sick of these highlevel tools like Python. Its the dumbest language ever... but everyone is addicted SOLELY for the ecosystem of tools and people are just lazy

You know what I hate? Videos that could have been articles.

Jason, or should I say json... you always pop up with videos right on the edge of what is useful, possible, and ahead. Thank you so much. Best AI practitioner on CZcams by some distance.

Regarding reasoning part, you could do the same with json output as well. At least, thats how i have been using json output.

So where you were using structured output wrong? Click bait? 😅

Hi Jason, many of us are just entering the world of AI and exploring its possibilities to make a living and support our families. However, not everyone can afford to subscribe to all the services available online. We follow several AI channels, but it's not feasible to pay for premium content from each one. Additionally, some users can only manage to pay for a subscription to services like ChatGPT or Claude. Please consider reevaluating your strategy and sharing more of your premium content with us. Thanks!

Thank you AI Jason

thanks, jason. This is a good introductory video about huggingface

Hi Jason! I was wondering what capabilities you think are possible with structured outputs compared to function calling? Your example of chain of though reasoning seems already possible using function calling. Generating UIs, which might use recursive (or nested), JSON definitions seems like an example that would require structured outputs. I'd love to hear your thought on this!

Incredible content dude. Thank you

Thanks Jin Yang.

Fantastic video, perfect pacing for me, with great real world examples. Thanks very much.

This is one of my favourite channels on AI. But many are good but this one great. I'm specifically looking at these models at the moment . TTS for sales in multiple languages and control of os and browser

great tutorial, thank you Jason! keep it going

Kind of a side question, but do we know presently which is the closes open source model that does that (pushing structured data) the the closest to the latest Chatgpt4? Hint: my test show that only gemma2 performs relatively well on structured output...

We need the llama or open source version of this tutorial

Great information Jason, thank you!

Quality as always

SO freaking good of the advanced tips, didnt realise you can push the reasoning like that

This is great!!

Get $50 AssemblyAI credits for free: www.assemblyai.com/? Also get full code breakdown & Join AI builder club: 2ly.link/1z94G

Awesomeeeee!!!!!! Strict JSON output is great!!

I have a doubt, since open ai api keys can only be used a certain amount of times for free. So , after deployment will people be able to still use it however long as they want ?

Hi I want to create a model which has cnversations like a specific person, I have a document (pdf), where this person is talking to another person , can i diirectly use this pdf instead of a csv file ?

Nice video, BTW do you have the new link for webQL ? Does not work

This is what were doing argh...

I want to train an LLM using videos and pictures. How can I achieve this? How do I make sure it can interpret the videos and pics?

and pip install not recognised problem coming

bro what's the simulator ?

Please drop link to the Google colab notebook

Why do you use two different ways to load env? load_dotenv and os.getenv. Why not just one way? Which way is better?

I like tool calling

Chat RTX has this program that will run a your machine and RAG a whole drive. Download.

I mostly download videos but after watching this video i had to come back to give it a comment, like , subscription and just give you a watch time. not time wasted .Thank you

Always bringing top notch content Jason, I really appreciate it, keep it up!

Llama3 integrated with local agentic RAG to leverage private knowledge sounds powerful! The potential for enhanced performance is exciting. #AI #Llama3 #Innovation

Valeu!

There is no more llama index right ?

What if I want to recreate character AI locally?