VMamba: Visual State Space Models，code is based on mamba

a Hassle-Free Python Experience

A high-throughput and memory-efficient inference and serving engine for LLMs

Robust recipes to align language models with human and AI preferences

This repository contains demos I made with the Transformers library by HuggingFace.

Dataframes powered by a multithreaded, vectorized query engine, written in Rust

Blazingly fast LLM inference.

☄🌌️ The minimal, blazing-fast, and infinitely customizable prompt for any shell!

Implementation for "Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs"

Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.

Ray is a unified framework for scaling AI and Python applications. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

Some basic examples of playing with RL

OpenVINO™ is an open-source toolkit for optimizing and deploying AI inference

Official repo for paper DigiRL: Training In-The-Wild Device-Control Agents with Autonomous Reinforcement Learning.

Development repository for the Triton language and compiler

[ICML2024 (Oral)] Official PyTorch implementation of DoRA: Weight-Decomposed Low-Rank Adaptation

A tiny scalar-valued autograd engine and a neural net library on top of it with PyTorch-like API

General technology for enabling AI capabilities w/ LLMs and MLLMs

4M: Massively Multimodal Masked Modeling

LLM101n: Let's build a Storyteller

A curated list of visual reinforcement learning resources

A minimal PyTorch re-implementation of the OpenAI GPT (Generative Pretrained Transformer) training

Repository for Meta Chameleon, a mixed-modal early-fusion foundation model from FAIR.

Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch

An Easy-to-use, Scalable and High-performance RLHF Framework (70B+ PPO Full Tuning & Iterative DPO & LoRA & Mixtral)

Train transformer language models with reinforcement learning.

OpenVLA: An open-source vision-language-action model for robotic manipulation.