Skip to content

CSGHub is an open-source large model platform just like on-premise version of Hugging Face. You can easily manage models and datasets, deploy model applications and setup model finetune or inference jobs with user interface. CSGHub also provides Python SDK with full compatibility of hf sdk. Join us together to build a safer and more open platform⭐️

License

Notifications You must be signed in to change notification settings

OpenCSGs/csghub

Repository files navigation

Unit Tests

Join WeChat Slack Join OpenCSG Slack Join OpenCSG Discord Gurubase

简体中文日本語한국어

CSGHub

CSGHub is an open-source platform designed for managing Large Language Models (LLM) assets, offering an efficient way to handle the entire LLM and their assets such as datasets, spaces and codes etc. It allows users to upload, download, store, verify, and distribute LLM assets via a web interface, git command line, a natural language Chatbot, or CSGHub SDK.

Additionally, CSGHub features microservice submodules and standardized OpenAPIs for seamless integration with existing systems. It aims to provide a user-friendly management platform specifically for LLM, which can be deployed On-Premise for secure, offline operation. Essentially, CSGHub serves as a private, on-premise version of Huggingface.

CSGHub

Key Features

  • Unified LLMs Management
  • Extensible Development Framework
  • Advanced Model Capabilities and Optimization
  • Space and Asset Management Assistant (Copilot)
  • Multi-Source Data Synchronization and Recommendation
  • Enterprise-Level Security and Access Control
  • On-Premises Deployment Solutions
  • E2E Data Processing and Intelligent Annotation System
  • Resilient High-Availability Architecture

For further information, please see the detailed introduction of CSGHub.

Demo Video

To help users get up to speed with CSGHub, we have created a demo video that highlights its key features and how it works. You can watch it below or on YouTube/Bilibili.

CSGHub_EN.mp4

Quick Start

  • For those looking to quickly explore, experiment with CSGHub's free SaaS version on the OpenCSG website. Refer to this brief quick start guide to handle LLMs/datasets and deploy LLM applications with CSGHub SaaS interface. For more information, please refer to the OpenCSG Documentation Center.

  • For those seeking a hands-on experience, you can quickly launch CSGHub on your local environment (Linux/MacOS/Windows) using the Docker one-click deployment guide, which allows you to start an all-in-one container with simplified commands and instantly experience the core functionalities of CSGHub.

  • For those deploying on Kubernetes, refer to the Helm Chart installation guide for step-by-step instructions.

Quick Try

one-click deployment on Ali marketplace

Roadmap

Discover the roadmap outlining CSGHub's future direction.

Release Notes

Refer to release notes for information on feature improvements.

Meetings & Contributing

We warmly invite you to explore the OpenCSG Community, where you'll find comprehensive guides on becoming a vital part of the project.

  • The process for joining community meetings, where you can engage with fellow members and share your ideas.
  • Refer to the contribution guide for strategies on how to contribute to the project.
  • Contact us for any questions or issues, ensuring you're always supported throughout your journey with OpenCSG.
  • Refer to the development guide for instructions on setting up the development environment.

About

CSGHub is an open-source large model platform just like on-premise version of Hugging Face. You can easily manage models and datasets, deploy model applications and setup model finetune or inference jobs with user interface. CSGHub also provides Python SDK with full compatibility of hf sdk. Join us together to build a safer and more open platform⭐️

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published