When we first started experimenting with local LLMs, it was a completely different story! We were using gaming GPUs to tinker around. 8GB or 16GB of VRAM (which wasn't even a given for everyone) was the norm, and so many people could actually get their hands dirty and experiment. Let’s just forget for a second that long crypto-mining phase that bloated the market and caused shortages... but today? Today, if you don't have high-end hardware, experimenting has become way too difficult. I know ...