TricorderO11y 最近的时间轴更新

llm-numbers-cn 中文版 llm-numbers

3k 3-k platform is for training LLMs

sxwl.ai SXWL.AI site

docs.sxwl.ai 算想未来产品文档站点

blog.sxwl.ai SXWL.AI's Blog Site

CaptainsLog SXWL.AI's significant operational records and roadmap

chat-ui

ChatGLM-6B ChatGLM-6B: An Open Bilingual Dialogue Language Model | 开源双语对话语言模型

ChatGLM-Efficient-Tuning Fine-tuning ChatGLM-6B with PEFT | 基于 PEFT 的高效 ChatGLM 微调

CMMLU CMMLU: Measuring massive multitask language understanding in Chinese

ConfigFiles Config files used for setting up development environment

deep-learning-pytorch-huggingface

DeepSpeedExamples Example models using DeepSpeed

dockerfiles Deep Learning Dockerfiles

goctl modified goctl for 3k project

gpu-operator NVIDIA GPU Operator creates/configures/manages GPUs atop Kubernetes

kserve Standardized Serverless ML Inference Platform on Kubernetes

LLaMA LLaMA fine tune

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

nccl Optimized primitives for collective multi-GPU communication

network-operator Mellanox Network Operator

node-feature-discovery Node feature discovery for Kubernetes

P-tuning-v2 An optimized deep prompt tuning strategy comparable to fine-tuning across scales and tasks

TricorderO11y 大模型算力云服务提供商 🏢  NascentCore.ai / CEO V2EX 第 584334 号会员，加入于 2022-06-10 13:11:36 +08:00

算想未来（https://nascentcore.ai）为中国 AI 经济打造世界领先的 AI IaaS Cloud，为中国大模型和 AI 应用开发者提供世界领先的基础设施软件服务。基于国产 GPU 的全新技术体系、构建独立自主可控的 AI IaaS Cloud，立足中国、服务全球大模型及 AI 应用开发者。