I turned years of local AI benchmarks into an app that helps you figure out what hardware actually makes sense for the models you want to run.
๐น Try Replit:
๐ Gear Links ๐
๐ ๏ธ๐ WisdPi 10G Ethernet Expansion Card for Framework:
๐ช๐ชHighly rated precision driver kit:
๐ปโ Favorite 15″ display with magnet:
๐งโก Great 40Gbps T4 enclosure:
๐ ๏ธ๐ My nvme ssd:
๐ฆ๐ฎ My gear:
๐ฅ Related Videos ๐ฅ
๐ธ๐ง Quant comparison vid – Everything looks fine at 4-bit
๐งณ๐งฐ Mini PC portable setup –
๐๐ป Dev setup on Mac –
๐ธ๐ง Cheap mini runs a 70B LLM ๐คฏ –
๐งช๐ฅ RAM torture test on Mac –
๐โก FREE Local LLMs on Apple Silicon | FAST! –
๐ง ๐ REALITY vs Appleโs Memory Claims | vs RTX4090m –
๐งฌ๐ Set up Conda –
โก๐ฅ Thunderbolt 5 BREAKS Appleโs Upcharge –
๐ง ๐ INSANE Machine Learning on Neural Engine –
๐งฑ๐ฅ๏ธ Mac Mini Cluster –
* ๐ ๏ธ Developer productivity Playlist –
๐ AI for Coding Playlist: ๐ –
Local AI Advisor: sorry, had to remove the link temporarily – lots of people โcontributingโ. itโs pretty funny, but not super helpful. weโll be back soon
โ โ โ โ โ โ โ โ โ
โค๏ธ SUBSCRIBE TO MY YOUTUBE CHANNEL ๐บ
Click here to subscribe:
โ โ โ โ โ โ โ โ โ
Join this channel to get access to perks:
โ โ โ โ โ โ โ โ โ
๐ฑ ALEX on X:
#replit #llm #ai
The only thing useful in this video is the repo and the link is off thanks Alex
Iโm looking to get into coding and some Ai. Iโve got a 16โ M4 24/512. I think this should be good enough for looking into what you are advertising. I forgot the name. Regit?
This is an interesting video and targeting something I’d really love to learn more about but I can’t use your app site that you built. It’s full of nonsense data now. So I feel like the point of the video is now completely lost until that’s addressed and data is cleaned. If that happens, I will turn my dislike into a like on this video, as it would be genuinely helpful.
“Dayter” please, not “datter” ๐
Make a video with the extreme cases like Kimmi K3 so ppl understand how crazy it gets (ofcourse not that far but anything that need a server would be cool)