Video

Is openai the only api? Or can i use local llm to mimik openai api will it work?

This is never going to work, since the LLM has to work from an OS as well. This is an OS on an OS. And it always needs lots of power.

Is there any way i can use these assistants created by phidata in multi agentic frameworks like crewai or autogen?

Do you have any playlist to learn concepts before this? I don't understand this code, I need more context

super cool, I hope the next CPUs will run Llama 3 70B fast

i have a question, If i try with Ollama llama3 locally i not sure this one was not working did you also face same issue >??

Legit question, why not just use the embed models like Nomic for example, chatting with my LLM I learned these vector "memories" create neural connection cells/nodes or whatever and it connects to these vector memories , meaning it's knowledge and memories sort of expands..

Kendi görüntün çok büyük. Kodları göremiyoruz.

Phidata > Praison AI?

This would work good with Enso the interactive programming language

Out of all the AI tools and frameworks you’ve used, which one(s) do you find to be the most useful and have the most promise moving forward?

How can I contact you? I would like to pay you for your assistance in regard to a project. Thank you so much!

You could have shared a link to the original video: czcams.com/video/6g2KLvwHZlU/video.html instead of recording a clone yourself.

Hi Mervin, what about postgres memory ? is it a long term memory ? something like autogen teachable agent or memgpt ? thx again for your amazing content !

❤❤❤

To clarify the code only works on mac?

do I have to do ollama run nomic-embed-text?

How to build a RAG application against a SQL Server database ?

why not use codelama ?

Physicists are trying to figure out why language models work intuitively why their reasoning capability emerges when you train them on language. And it comes out that the universe is upheld by the word of God. We were designed in a sense as a language model to speak what we see and observe. Of course, our peak creation would do the same. Gpt agrees with me. Also, the Bible itself was made for a language model. It's a keyword indexed database. Ready for consumption . I hope more believers who understand the grace of Christ will start building rag systems to educate people from their perspective. Unfortunately, so many are afraid of AI. They're missing this window of opportunity which may be short-lived who knows

Yay. This is a unique opportunity to handle the word in a way that we never have before.

can i ask for that json data?

Can this be used to extract specific data fields from multiple pdf files and output the result in Excel?

Awesome tutorial Mervin! Thanks!

Do we need to provide Groq Api key or just keep as it is here.

excellent

Autogen or taskweaver which one?

cant we use python not docker its very ram consuming

Great video! I have a question here. Do you have any idea on creating an agent that can analyze the legal contracts uploaded by user? Example tasks are: 1. Identify the parties in the contract 2. Identify and extract all the clauses in the contract 3. Summarize the important details in each identified clause 4. Draft potential missing clauses or clauses can be added into the current contract 5. Draft the whole legal document if possible (Optional bcus this is too difficult I think)

Can I use firebase instead of gcloud

nice content!

Hi does this work for fine-tuned local llama 3 model?

Ohh okadarda akili degilmiş backler hala sağlam

People need to connect OpenDevin to Enso interactive programming language, this would speed up development extremely quick.

Same

doesn't perplexity ai do this already?

This is complete BS, it doesn't actually work. It takes ages to learn how to things on your system.

Imagine integrating this with GPT4o having a real AI interview instead of typing

Wonderful, but shocking. But wonderful. I‘m looking for a way to catch a call from my iPhone (hopefully on my local mac) to then answered by a local ai or groq. I have no idea where to start.

You seem to be very active in this field and produce a lot of what appears to be helpful videos. If only a non-developer could follow them successfully. I have tried a few of your tutorials and must conclude that I have to plan on spending a significant amount of time trying to decipher them and "translate" them into a windows environment and after all of that I can be certain to not be successful. There is always something that does not work or something that is missing or not explained or the audio comes across mumbled and I ask myself WTF. I understand you are so good and so quick at what you're doing, but please slow it down a bit and don't presume we all understand the fine little in-between-steps that are missing from your tutorials. Trying hard but frustrated.

I feel compelled to say that I see this as .. just .. existentially wrong. We already have recruitment agents and HR departments that are basically scanning and filtering CVs/resumes for/on keywords without real knowledge and understanding of the skillset required. Who checks the correctness of the LLMs questions, answers? Who certifies that the LLM has no biases? - how do they do that? - to a standard that will hold-up in court whenever the inevitable legal action ensues? Presumably year are only talking initial filtering and technical interview stages here and not team-fit and soft skills but what happens if soneone gives an inspired but non-textbook answer (I'm thinking fast inverse square root). And ... if the AI is good enough in a field to actually conduct a valid, thorough and useful interview then .. for a lot of jobs at least .. why is the company hiring for the roke and not just using the AI to do it? In an IT context: junior devs don't conduct interviews. If the AI is at the level of a good senior dev then .. its good enough to do the job so the interview is mostly pointless(?)

It'll be great if you could explore doing similar things using n8n also, which can orchestrate all kinds of workflows

This is Amazing! <3

czcams.com/video/qaDWwOpjV3M/video.htmlsi=4Gv1d-JUyuqrsevd

can we try agentic chunking with ollama?

When is it coming to India?

Your a god, for me. My next steps will be to create agents for find budgets for researchers, with french administrations.

@mervin would you be able to loop the crew to simulate a multiagent chat ui?

Need a new video for assistant api with new features.I need to know how to control token count?????

Good explanation. But it is impossible to read line 185 and by consequence, impossible to test the best snippet that you are soposed to teaching