💗 Thank you so much for watching guys! I would highly appreciate it if you subscribe (turn on notifcation bell), like, and comment what else you want to see! 📆 Book a 1-On-1 Consulting Call WIth Me: calendly.com/worldzofai/ai-consulting-call-1 🔥 Become a Patron (Private Discord): patreon.com/WorldofAi 📖 Want to Hire Me For AI Projects? Fill Out This Form: td730kenue7.typeform.com/to/WndMD5l7 🚨 Subscribe to my NEW Channel! www.youtube.com/@worldzofcrypto 🧠 Follow me on Twitter: twitter.com/intheworldofai Love y'all and have an amazing day fellas. ☕To help and Support me, Buy a Coffee or Donate to Support the Channel: ko-fi.com/worldofai - Thank you so much guys! Love yall!
@intheworldofai7 ай бұрын
[MUST WATCH]: OpenDevin: AI Software Engineer With Complex Coding Completion: kzbin.info/www/bejne/g2S3hoWBfpWDprssi=aOtFx9erOS8V9DBG Open Interpreter's 01 Lite: Open-Source Personal AI Agent!: kzbin.info/www/bejne/p2HHe2qKbMaoaLssi=flIwTvbLRBmH7_9k Devika: Opensource AI Software Engineer! Builds & Deploy Apps End-to-End!: kzbin.info/www/bejne/fXPZf6t-ntaoga8si=CoyMlfQtWLBiGUDl
@intheworldofai6 ай бұрын
aiXcoder 7B: Powerful Coding LLM for Developers - Writes Code For You!: kzbin.info/www/bejne/imPNgIOtpc2Dmdk
@Raubritterr2227 ай бұрын
what is that activity monitor app in terminal in the beginning of the video?
@simonhill626711 күн бұрын
whats the difference in reality between the Instruct 3b model versus the normal 3B model ?
Qwen1.5 MoE: Powerful Mixture of Experts Model - On Par with Mixtral! kzbin.info/www/bejne/oXa2eYJpmJiElas
@unknown.45957 ай бұрын
thanks for video
@Alex-gc2vo7 ай бұрын
Why would anyone want a 3B model? Anything a 3B can do a 7B can do better, and 7B runs perfectly fine on all hardware
@id1043354097 ай бұрын
All models work with Python. Why?
@awesomebearaudiobooks7 ай бұрын
Because Python is the language that usually has the shortest, more concise code when compared to other common languages. And AI has both a limited parameter count and a limited context window, so it is significantly easier to both train and to use AI with Python than with almost anything else. Perhaps there are "even more concise" languages to teach to AI, like Elixir or Closure, but they are way less common, and also they are not used for AI development, so that AI companies can't use these languages to improve AI even faster. So they stick with mostly training in Python.
@jeffg46867 ай бұрын
under 2 gb, nice - any machine
@leroymason27627 ай бұрын
Still haven't found any small llm, thats better than deep seek coder.
@Leonhart_937 ай бұрын
I am not sure anything can beat deepseek coder 33B. It's very good.
@s6yx7 ай бұрын
have a hf link to it?
@GenAIWithNandakishor7 ай бұрын
Deep seek coder is a beast
@fredoscott23467 ай бұрын
Try to build something useful...
@Leonhart_937 ай бұрын
Unfortunately there is no way a 3B model can ever beat DeepSeek Coder 33B, which has 11x as many parameters and all of them are from computer science data. Even if stuff is made more efficient, there is no way to overcome the data disparity.
@AmberLois7 ай бұрын
Is DeepSeek Coder better than GPT-4 in coding? Like can it solve more problems EASILY than GPT-4?
@ПетрФомин-щ9ж7 ай бұрын
Unfortunately all models smaller than 40 gigabytes are just useless and the model in the video is hardly an exception.
@lendo14577 ай бұрын
Well... I dont think you really get llms
@GenAIWithNandakishor7 ай бұрын
They need to really work on EOS token. at the end is annoying