In this video, I dive into the world of multimodal Large Language Models (LLMs) with a focus on two powerful models: MiniCPM (8B parameters) and LLaVA (13B parameters). I start by explaining how multimodal LLMs work and explore some of their key use cases in various industries, including construction.
I then introduce the system I’m building, structured around three key pillars:
1. Ollama - A versatile platform for deploying LLMs.
2. Docker Desktop - A containerization tool that simplifies application deployment.
3. Open WebUI - A user-friendly interface for running different LLMs.
In the video, I walk you through the step-by-step installation of each component. First, I show you how to install Ollama on Linux, then cover how to set up Docker Desktop on Linux by installing Docker Engine. Next, I explain how to install Open WebUI, including how to load models from HuggingFace and Ollama, and how to start a chat with various LLMs.
In the final part of the video, I showcase the deployment of the entire system. I demonstrate the performance of both multimodal LLMs in interpreting text from images and understanding the image context. I also explore some useful applications within the construction industry.
This system is low-code to no-code, making it easy for anyone to deploy without any coding experience. Plus, it’s free to use!
🔍 Agenda:
⚫00:13 Intro
⚫00:52 Use Cases
⚫01:34 Installing Ollama
⚫03:22 Installing Docker
⚫07:17 Installing Open WebUI
⚫12:39 System Deployment
⚫19:38 Outro
👨💻 Whether you're an LLM developer or an APC candidate, this video breaks down the entire process from dataset conversion to model deployment, helping you achieve accurate, fine-tuned results.
🌐 Visit www.apcmasterypath.co.uk for tailored packages in RICS APC preparation and expert advice on integrating AI into the construction industry with practical, deliverable solutions.
👍 Don’t forget to subscribe to APC Mastery Path on RU-vid for more insights, tips, and tutorials on mastering your APC journey!
🔔 Don’t forget to like, subscribe, and hit the notification bell to stay updated with the latest tutorials and insights into AI, APC exam preparation, and LLM development!
🔗 General Links & Resources:
⚫Our Website: www.apcmasterypath.co.uk
⚫All APC Mastery Path Blogposts: www.apcmastery...
⚫Personal Linkedin Page: / mohamed-ashour-0727
⚫APC Mastery Path Linkedin Page: / apc-mastery-path
📽️Useful videos:
⚫ Unsloth FineTuning & Comparing LLMs:Mistral, Gemma 2, Llama 3.1 with Chatbot Deployment on OpenWebUI: • Unsloth FineTuning & C...
⚫Create Training Data for Finetuning LLMs: • Create Training Data f...
⚫Finetune your LLMs on custom datasets using Unsloth: • Finetune Your LLM on C...
⚫Deploy Open WebUI with Zero Coding Skills : • Unlocking Local AI: De...
📝Prerequisites & Dependencies:
⚫Python for Windows : www.python.org...
⚫Python for Linux: docs.python-gu...
⚫Nvidia Cuda Toolkit v 12.1: developer.nvid...
⚫ Install Linux on your windows PC : www.minitool.c...
⚫Windows subsystem for Linux : learn.microsof...
⚫Anaconda for Linux: repo.anaconda....
⚫ Pytorch: pytorch.org/
⚫Ollama : www.ollama.com/download
⚫Docker for Windows: desktop.docker...
⚫Docker for Linux: docs.docker.co...
⚫Open WebUI on Github: github.com/ope...
⚫Visual Studio Code: code.visualstu...
📚Github & Huggingface repositories:
⚫Unsloth Huggingface repository: huggingface.co...
⚫Unsloth GitHub Repository: github.com/uns...
#MultimodalLLMs #MiniCPM #LLaVA #Ollama #Docker #OpenWebUI #LowCode #NoCode #AIinConstruction #AIinConstruction #LLMFineTuning #RICSAPC #Unsloth #Mistral7b #Gemma2 #Llama31 #AIMastery #APCMasteryPath #OpenWebUI #APCTraining #Llama3.1 #Unsloth #LLMFineTuning #OpenWebUI #Huggingface #RICSAPC #BoQCodification #NRM1 #ChatbotDevelopment #AITraining #QuantitySurveying
2 окт 2024
