Code DeepSeek V3 From Scratch in Python - Full Course

Published: 01 April 2025
on channel: freeCodeCamp.org

105,217

3.4k

This course is a comprehensive guide to understanding and implementing DeepSeek V3, a cutting-edge deep learning model. ‪@vukrosic‬ shares step-by-step coding instructions and theoretical insights.

🔗 paper - https://arxiv.org/pdf/2412.19437

💻 https://github.com/deepseek-ai/DeepSe... - code by DeepSeek, just a few small changes made at the end of the video to Transformer class to for training, as this is for inference, so you need to make them manually or screenshot the video and ask AI to make the changes to this code

❤️ Try interactive AI courses we love, right in your browser: https://scrimba.com/freeCodeCamp-AI (Made possible by a grant from our friends at Scrimba)

⭐️ Contents ⭐️
⌨️ (0:00:00) Intro
⌨️ (0:01:40) Attention Mechanism
⌨️ (0:13:34) Query, Key, Value
⌨️ (0:34:11) KV Cache
⌨️ (0:39:06) Multihead Latent Attention (MLA)
⌨️ (0:58:53) Coding MLA
⌨️ (1:28:41) RoPE
⌨️ (1:55:44) Coding KV Cache
⌨️ (2:00:25) MLA forward
⌨️ (2:28:24) MoE, Gate
⌨️ (2:49:25) Gate code
⌨️ (3:09:10) MoE code
⌨️ (3:28:36) Transformer Blocks

🎉 Thanks to our Champion and Sponsor supporters:
👾 Drake Milly
👾 Ulises Moralez
👾 Goddard Tan
👾 David MG
👾 Matthew Springman
👾 Claudio
👾 Oscar R.
👾 jedi-or-sith
👾 Nattira Maneerat
👾 Justin Hual

--

Learn to code for free and get a developer job: https://www.freecodecamp.org

Read hundreds of articles on programming: https://freecodecamp.org/news

On this page of the site you can watch the video online Code DeepSeek V3 From Scratch in Python - Full Course with a duration of hours minute second in good quality, which was uploaded by the user freeCodeCamp.org 01 April 2025, share the link with friends and acquaintances, this video has already been watched 105,217 times on youtube and it was liked by 3.4 thousand viewers. Enjoy your viewing!

168