mzchtx

AISystem

AISystem 主要是指AI系统，包括AI芯片、AI编译器、AI推理和训练框架等AI全栈底层技术

nanoGPT

The simplest, fastest repository for training/finetuning medium-sized GPTs.

unsloth

Finetune Llama 3.1, Mistral, Phi & Gemma LLMs 2-5x faster with 80% less memory

skypilot

SkyPilot: Run AI and batch jobs on any infra (Kubernetes or 12+ clouds). Get unified execution, cost savings, and high GPU availability via a simple interface.

GLM-4

GLM-4 series: Open Multilingual Multimodal Chat LMs | 开源多语言多模态对话模型

simple-evals

veScale

A PyTorch Native LLM Training Framework

llama3-from-scratch

llama3 implementation one matrix multiplication at a time

LLaMA-Factory

Efficiently Fine-Tune 100+ LLMs in WebUI (ACL 2024)

Megatron-LM

Ongoing research training transformer models at scale

dlrover

DLRover: An Automatic Distributed Deep Learning System

Awesome-LLM-Compression

Awesome LLM compression research papers and tools.

unilm

Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities

LLM-Viewer

Analyze the inference of Large Language Models (LLMs). Analyze aspects like computation, storage, transmission, and hardware roofline model in a user-friendly interface.

Qwen2

Qwen2 is the large language model series developed by Qwen team, Alibaba Cloud.

grok-1

Grok open release

optimum-nvidia

SoraReview

The official GitHub page for the review paper "Sora: A Review on Background, Technology, Limitations, and Opportunities of Large Vision Models".

llama2.c

Inference Llama 2 in one file of pure C

ragas

Evaluation framework for your Retrieval Augmented Generation (RAG) pipelines

FlagEmbedding

Retrieval and Retrieval-augmented LLMs

EAGLE

Official Implementation of EAGLE-1 and EAGLE-2

Yi

A series of large language models trained from scratch by developers @01-ai

tabby

Self-hosted AI coding assistant

LLMLingua

To speed up LLMs' inference and enhance LLM's perceive of key information, compress the prompt and KV-Cache, which achieves up to 20x compression with minimal performance loss.

lorax

Multi-LoRA inference server that scales to 1000s of fine-tuned LLMs

PhotoMaker

PhotoMaker [CVPR 2024]

mergekit

Tools for merging pretrained large language models.

TigerBot

TigerBot: A multi-language multi-task LLM

RAG-Survey