🚀 Set Up Ollama with Tesla K80 – Full Guide for CUDA 11.4 + GCC 10 + Go + CMake 💻
In this video, I’ll walk you step-by-step through setting up Ollama to run on an NVIDIA Tesla K80 GPU using CUDA 11.4. We’ll handle all the heavy lifting — from installing the NVIDIA Toolkit, compiling GCC 10 and CMake from source, and setting up Go (Golang) using the official binary.
🛠️ What You'll Learn:
Installing the correct NVIDIA CUDA Toolkit for K80
Building GCC 10 from source to meet compatibility needs
Compiling CMake manually for maximum control
Installing Go from binary for a faster setup
Prepping your system for Ollama and LLM dev workflows
🔧 Whether you're setting up a dev box, repurposing old hardware, or just want to explore LLMs with a Tesla K80, this guide has got you covered.