DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

SGLang is a structured generation language designed for large language models (LLMs). It makes your interaction with models faster and more controllable.

Xtuner Factory

XTuner is a toolkit for efficiently fine-tuning LLM

OpenCompass is an LLM evaluation platform, supporting a wide range of models (Llama3, Mistral, InternLM2,GPT-4,LLaMa2, Qwen,GLM, Claude, etc) over 100+ datasets.

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper and Ada GPUs, to provide better performance with lower memory utilizatio…

Inference code for CodeLlama models

An efficient, flexible and full-featured toolkit for fine-tuning LLM (InternLM2, Llama3, Phi3, Qwen, Mistral, ...)

🤗 Diffusers: State-of-the-art diffusion models for image and audio generation in PyTorch and FLAX.

[ICLR'24 spotlight] An open platform for training, serving, and evaluating large language model for tool learning.

Home of StarCoder: fine-tuning & inference!

An open-source tool-augmented conversational language model from Fudan University

LMDeploy is a toolkit for compressing, deploying, and serving LLMs.

Scripts for fine-tuning Meta Llama3 with composable FSDP & PEFT methods to cover single/multi-node GPUs. Supports default & custom datasets for applications such as summarization and Q&A. Supportin…

The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

Development repository for the Triton language and compiler

[ICCV 2023] OpenOccupancy: A Large Scale Benchmark for Surrounding Semantic Occupancy Perception

OpenMMLab YOLO series toolbox and benchmark. Implemented RTMDet, RTMDet-Rotated,YOLOv5, YOLOv6, YOLOv7, YOLOv8,YOLOX, PPYOLOE, etc.

Model Quantization Benchmark

OpenMMLab Model Compression Toolbox and Benchmark.

OpenMMLab Foundational Library for Training Deep Learning Models

PaddleSlim is an open-source library for deep model compression and architecture search.

This is a knowledge distillation toolbox based on mmdetection.

OpenMMLab Model Deployment Framework

OpenMMLab Pre-training Toolbox and Benchmark

OpenMMLab FewShot Learning Toolbox and Benchmark

OpenMMLab Detection Toolbox and Benchmark

OpenMMLab Semantic Segmentation Toolbox and Benchmark.