《大模型白盒子构建指南》：一个全手搓的Tiny-Universe

🚀 「大模型」1小时从0训练26M参数的视觉多模态VLM！🌏 Train a 26M-parameter VLM from scratch in just 1 hours!

🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!

📚LeetCUDA: Modern CUDA Learn Notes with PyTorch for Beginners🐑, 200+ CUDA Kernels, Tensor Cores, HGEMM, FA-2 MMA.🎉

kindle电子书

个人构建MoE大模型：从预训练到DPO的完整实践

Performance analysis of predictive (alpha) stock factors

We are committed to the open-sourcing quantitative knowledge, aiming to bridge the information gap between the domestic and international quantitative finance industries. 我们致力于量化知识的开源与汉化，打破国内外量化金融行…

An intuitive library to extract features from time series. To cite this software publication: https://www.sciencedirect.com/science/article/pii/S2352711020300017

Muon is an optimizer for hidden layers in neural networks

My learning notes for ML SYS.

✔（已完结）最全面的 深度学习 笔记【土堆 Pytorch】【李沐 动手学深度学习】【吴恩达 深度学习】【大飞 大模型Agent】

slime is an LLM post-training framework for RL Scaling.

Muon is Scalable for LLM Training

EasyRL: An easy-to-use and comprehensive reinforcement learning package.

[ICLR 2025] 🧬 RegMix: Data Mixture as Regression for Language Model Pre-training (Spotlight)

SGLang is a fast serving framework for large language models and vision language models.

A collection of multimodal reasoning papers, codes, datasets, benchmarks and resources.

✨✨Latest Advances on Multimodal Large Language Models

Reproduce R1 Zero on Logic Puzzle

R1-searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning

Python语言基础50课

FinRobot: An Open-Source AI Agent Platform for Financial Analysis using LLMs 🚀 🚀 🚀

A course of learning LLM inference serving on Apple Silicon for systems engineers: build a tiny vLLM + Qwen.

Puzzles for learning Triton