The full minitorch student suite.

Accelerate inference without tears

slime is a LLM post-training framework aiming at scaling RL.

A lightweight reinforcement learning framework that integrates seamlessly into your codebase, empowering developers to focus on algorithms with minimal intrusion.

Democratizing Reinforcement Learning for LLMs

verl: Volcano Engine Reinforcement Learning for LLMs

A high-performance distributed deep learning system targeting large-scale and automated distributed training.

Tiny-FSDP, a minimalistic re-implementation of the PyTorch FSDP

Ring attention implementation with flash attention

Nano vLLM

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

Scalable toolkit for efficient model reinforcement

一个简单的恰到好处LLM应用框架，能够让你以最“Code Center“的方式无缝集成LLM能力

TextPy: Collaborative Agent Workflow through Programming and Prompting

My learning notes/codes for ML SYS.