Build Your Own ChatGPT with LangChain & Ollama (No OpenAI API Key Needed!)

Want to build your own ChatGPT-like chatbot without relying on OpenAI's API? This video shows you how to do just that using LangChain and Ollama! LangChain is a powerful framework for building AI-powered applications, and Ollama is the latest cutting-edge platform to run large language model that can generate human-quality text.
Join us as we dive into the world of AI and build this chatbot using typescript/javascript.
By the end of this video, you'll have your very own AI assistant ready to answer your questions and complete your requests!
Github Repo: github.com/Kou...
Quickstart Guide to Ollama: • Ollama.ai: A Developer...
Stream LLM Responses in NextJS chat app: • Build a Next.js App wi...
Make Ollama WebAPI Public: • Host a AI Server

[@shreyas.sihasane]

Great work, please bring more AI projects

[@Aman_yadav1419]

Amazing keep uploading ❤

[@haashirnawaz9687]

the streamingtextresponse is not working anymore from the ai library what do i do instead to make it work?

[@MrRobots100]

Few questions: 1) Will it work locally on the device on which it is installed on for example lets say react native or do you need to set up a flask API and get response 2) If I have a MongoDb or Firebase DB with data and I want to finetune the model to give me responses based on that, can I do it 3) Scalability costs, for lets say from 1000 users to 100,000 users, how can I keep it low

[@maplearcadecode]

1) No, you don't need to set up flask API for this. Installing Ollama will spin up a web api. Do check out my Ollama quick start guide and you'll have your question answered. 2) Of course, you can load fine tuned custom models on ollama. You can check out this tutorial here (kzbin.info/www/bejne/aXPRn6aHeNKagtEsi=P4yZYxFuGtyKmmU6) 3) If you are thinking about sharing the app then as a first step, check out ngrok and their pricing to make the locally hosted app public. Secondly, the shared app will be limited by the throughput of the LLM itself, which will be influenced by the local hardware you will be running it on. You might wanna consider running ollama in hosted environments within a docker.

[@deepanshjha6353]

Could you please provide guidance on creating an Advanced RAG model that incorporates Hugging Face LLM models and PDFs within Next.js using TypeScript?

[@maplearcadecode]

Sure I'll make a video soon

[@Pancelet]

@@maplearcadecode faster please

[@hasanulhaquebanna]

How can we train this model or how can built our own ai model to work with our needs?

[@ash_b9875]

Thank you very much