Highlights
- Pro
Stars
(best/better) practices of megatron on veRL and tuning guide
Train speculative decoding models effortlessly and port them smoothly to SGLang serving.
slime is an LLM post-training framework for RL Scaling.
A toolkit for reproducible reinforcement learning research.
Genai-bench is a powerful benchmark tool designed for comprehensive token-level performance evaluation of large language model (LLM) serving systems.
📚LeetCUDA: Modern CUDA Learn Notes with PyTorch for Beginners🐑, 200+ CUDA Kernels, Tensor Cores, HGEMM, FA-2 MMA.🎉
Machine Learning Engineering Open Book
how to optimize some algorithm in cuda.
aider is AI pair programming in your terminal
Efficient Triton Kernels for LLM Training
Development repository for the Triton language and compiler
Examples and guides for using the OpenAI API
Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train OpenAI gpt-oss, DeepSeek-R1, Qwen3, Gemma 3, TTS 2x faster with 70% less VRAM.
Python clone of Spark, a MapReduce alike framework in Python
Scalable and flexible workflow orchestration platform that seamlessly unifies data, ML and analytics stacks.
Feathr – A scalable, unified data and AI engineering platform for enterprise
Kyle-Kyle / EverydayWechat
Forked from sfyc23/EverydayWechat每日自动给女朋友发微信暖心话。
刷算法全靠套路,认准 labuladong 就够了!English version supported! Crack LeetCode, not only how, but also why.
Python - 100天从新手到大师
💬 Easy way to create conversation chats
the t-rex runner game extracted from chromium
ReactJS version of Director Responsive Admin Template Free
Solutions to LeetCode problems; updated daily. Subscribe to my YouTube channel for more.
💥 A boilerplate application for building RESTful APIs Microservice in Node.js using express and mongoose in ES6 with code coverage and JsonWebToken Authentication
Script to generate verified Nike+ accounts using getsmscode.com