MindSpore
AI Framework for All Scenarios
Through community cooperation, this open AI framework best matches with Ascend processors and supports multi-processor architectures for all scenarios. It brings data scientists, algorithm engineers, and developers with friendly development, efficient running, and flexible deployment, and boosts the development of the AI software and hardware ecosystem.
Native Distributed Model Training
Provides multiple parallel capabilities for foundation model training, as well as easy-to-use APIs for configuring distributed policies.
AI4S Converged Computing Framework
Fully unleashing hardware potential for scientific computing with converged AI and simulation capabilities.
Quick Deployment in All Scenarios
Seamlessly deploy models across cloud, edge, and device scenarios with optimized runtime performance.
Knowledge Map
Mastering MindSpore: From Beginner to Advanced
Tutorials
From Quickstart to Experts
Video Courses
Online courses, from beginner to expert
MindSpore Activities
Community Communication and Sharing
Events & News
MindSpore 2.8 Released
HyperParallel Architecture Designed for Supernodes. Experience the next generation of AI framework performance.
Community Meetup
Join the MindSpore community events to share insights, learn best practices, and connect with fellow developers.
Developer Blog
Stay updated with the latest technical blogs, tutorials, and best practices from the MindSpore team.
Install
Installing MindSpore requires access to the public internet. If you are in an internal network environment, ensure that the network connection is properly configured.
Accessing a Cloud Platform
Cloud platforms help users quickly create and deploy models and manage the entire AI workflow. Choose a cloud platform to get started with MindSpore.
ModelArts
Huawei Cloud's full-stack AI platform providing data processing, model training, and deployment services.
OpenI
An open-source AI collaboration platform for community-driven model development and sharing.
Start Learning
Quick Start
An introduction to Huawei AI full stack is given and the implementation of common deep learning tasks is described through the basic MindSpore functions.
Image Classification
The CIFAR-10 dataset is used to describe how MindSpore processes image classification tasks.
Sentiment Analysis
Build an NLP model for text analysis and inference to analyze and classify sentiments.
MindSpore LLM Platform
Learn and practise
Repositories
Ecosystem
Provides open source AI research projects, case collections, and task-specific SOTA models and derivatives.
Go to view all projectsMindSpore Hub
A centralized repository for pre-trained models, allowing developers to discover, download, and deploy models.
Explore modelsMindSpore Serving
A lightweight and high-performance service module for deploying MindSpore models in production environments.
Learn moreContact and Help
AtomGit
Any bugs or advices, submit an Issue in AtomGit
Forum for Help
Post to discuss, get professional answers
Cooperative Association
Interested cooperation can be contacted by email
contact@public.mindspore.cnAchievements
Become part of the MindSpore community and help shape the future of AI development.
Community Partners
Collaborating with leading organizations to drive AI innovation forward.