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

Code and models for NExT-GPT: Any-to-Any Multimodal Large Language Model

很多镜像都在国外。比如 gcr 。国内下载很慢，需要加速。致力于提供连接全世界的稳定可靠安全的容器镜像服务。

Fast-Chat-X-Client-Python

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

Python bindings for llama.cpp

📚150+ Tensor/CUDA Cores Kernels, ⚡️flash-attn-mma, ⚡️hgemm with WMMA, MMA and CuTe (98%~100% TFLOPS of cuBLAS/FA2 🎉🎉).

learning how CUDA works

LLM training in simple, raw C/CUDA

通过浏览器渲染生成表格图像

本项目旨在收集开源的表格智能任务数据集（比如表格问答、表格-文本生成等），将原始数据整理为指令微调格式的数据并微调LLM，进而增强LLM对于表格数据的理解，最终构建出专门面向表格智能任务的大型语言模型。

User-friendly AI Interface (Supports Ollama, OpenAI API, ...)

Summarize existing representative LLMs text datasets.

Mini HoK: a novel MARL benchmark based on the popular mobile game, Honor of Kings, to address limitations in existing environments such as complexity and accessibility.

The most comprehensive database of Chinese poetry 🧶最全中华古诗词数据库, 唐宋两朝近一万四千古诗人, 接近5.5万首唐诗加26万宋诗. 两宋时期1564位词人，21050首词。

This repository contains datasets and baselines for benchmarking Chinese text recognition.

Generate text images for training deep learning ocr model

The official repo of Qwen-VL (通义千问-VL) chat & pretrained large vision language model proposed by Alibaba Cloud.

Personal Project: MPP-Qwen14B & MPP-Qwen-Next(Multimodal Pipeline Parallel based on Qwen-LM). Support [video/image/multi-image] {sft/conversations}. Don't let the poverty limit your imagination! Tr…

Unified Efficient Fine-Tuning of 100+ LLMs (ACL 2024)

AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation:

TensorRT-LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and build TensorRT engines that contain state-of-the-art optimizations to perform inference efficie…

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

Official Implementation of EAGLE-1 (ICML'24) and EAGLE-2 (EMNLP'24)

[ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding

Fast and low-memory attention layer written in CUDA

🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production

Fast and memory-efficient exact attention

用于从头预训练+SFT一个小参数量的中文LLaMa2的仓库；24G单卡即可运行得到一个具备简单中文问答能力的chat-llama2.