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

🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support

Accelerate local LLM inference and finetuning (LLaMA, Mistral, ChatGLM, Qwen, DeepSeek, Mixtral, Gemma, Phi, MiniCPM, Qwen-VL, MiniCPM-V, etc.) on Intel XPU (e.g., local PC with iGPU and NPU, discrete GPU such as Arc, Flex and Max); seamlessly integrate with llama.cpp, Ollama, HuggingFace, LangChain, LlamaIndex, vLLM, DeepSpeed, Axolotl, etc.

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

An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries

Example models using DeepSpeed

cube studio开源云原生一站式机器学习/深度学习/大模型AI平台，mlops算法链路全流程，算力租赁平台，notebook在线开发，拖拉拽任务流pipeline编排，多机多卡分布式训练，超参搜索，推理服务VGPU虚拟化，边缘计算，标注平台自动化标注，deepseek等大模型sft微调/奖励模型/强化学习训练，vllm/ollama/mindie大模型多机推理，私有知识库，AI模型市场，支持国产cpu/gpu/npu 昇腾生态，支持RDMA，支持pytorch/tf/mxnet/deepspeed/paddle/colossalai/horovod/ray/volcano等分布式

AllTalk is based on the Coqui TTS engine, similar to the Coqui_tts extension for Text generation webUI, however supports a variety of advanced features, such as a settings page, low VRAM support, DeepSpeed, narrator, model finetuning, custom models, wav file maintenance. It can also be used with 3rd Party software via JSON calls.

MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.

cube studio开源云原生一站式机器学习/深度学习/大模型AI平台/MaaS/mlops/人工智能平台/训推平台，算法全链路流程，算力租赁平台，拖拉拽任务流pipeline编排，多机多卡分布式训练，超参搜索，推理服务，VGPU虚拟化，云边端协同，边缘计算，自动化标注平台，deepseek等大模型sft微调/奖励模型/强化学习训练，vllm/ollama/mindie大模型多机推理，私有知识库llmops智能体，AI模型市场，支持国产异构算力调度,昇腾/寒武纪/海光/摩尔/沐曦等，支持ib/roce/RDMA，支持pytorch/deepspeed/colossalai/ray等分布式

Safe RLHF: Constrained Value Alignment via Safe Reinforcement Learning from Human Feedback

Ongoing research training transformer language models at scale, including: BERT & GPT-2

An Open-sourced Knowledgable Large Language Model Framework.

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! Train your own 8B/14B LLaVA-training-like MLLM on RTX3090/4090 24GB.

Best practices & guides on how to write distributed pytorch training code

GLake: optimizing GPU memory management and IO transmission.

Large Language Models for All, 🦙 Cult and More, Stay in touch !

Guide: Finetune GPT2-XL (1.5 Billion Parameters) and finetune GPT-NEO (2.7 B) on a single GPU with Huggingface Transformers using DeepSpeed

Collaborative Training of Large Language Models in an Efficient Way

chatglm多gpu用deepspeed和

Super-Efficient RLHF Training of LLMs with Parameter Reallocation

Deepspeed、LLM、Medical_Dialogue、医疗大模型、预训练、微调

Implementation of Chinese ChatGPT

No description available

一个基于HuggingFace开发的大语言模型训练、测试工具。支持各模型的webui、终端预测，低参数量及全参数模型训练(预训练、SFT、RM、PPO、DPO)和融合、量化。

train llama on a single A100 80G node using 🤗 transformers and 🚀 Deepspeed Pipeline Parallelism

中文大模型微调(LLM-SFT), 数学指令数据集MWP-Instruct, 支持模型(ChatGLM-6B, LLaMA, Bloom-7B, baichuan-7B), 支持(LoRA, QLoRA, DeepSpeed, UI, TensorboardX), 支持(微调, 推理, 测评, 接口)等.

DeepSpeed教程 & 示例注释 & 学习笔记 （大模型高效训练）

llama2 finetuning with deepspeed and lora

使用deepspeed从头开始训练一个LLM,经过pretrain和sft阶段,验证llm学习知识、理解语言、回答问题的能力