Build a future-ready AI framework to create infinite possibilities over device, edge, and cloud scenarios. Join us to unleash the power of AI with MindSpore.mindspore.cn/enJoined December 2021
Unexpected failures can disrupt long-running model training. With #MindSpore, training process graceful exit proactively saves checkpoints, supporting efficient training continuation while reducing resource waste.
⬇️ Make your training more resilient:
mindspore.cn/tutorials/en/r…
How can image restoration models handle unknown degradations?
With #MindSpore, the Out-of-Sample Restoration (OSR) task uses model reprogramming to translate unknown degradations into recognizable patterns for generalized restoration.
Discover the paper:
arxiv.org/abs/2403.05886
#AI is no longer limited to text. Training MLLMs introduces new challenges.
#MindSpore HyperParallel-Mpipe brings an algebra-driven approach to multimodal PP, organizing heterogeneous computations into efficient execution schedules.
🚀 Boost MLLM training:
atomgit.com/mindspore/hype…
🧐 Worried about costly failures during model training? Don't start from scratch! Learn how #MindSpore's fault recovery uses checkpoints to save parameters periodically, allowing quick recovery & seamless continuation.
⬇️ Keep your training on track! mindspore.cn/tutorials/en/r…
In model training, try .𝐚𝐠𝐞𝐧𝐭/ for higher efficiency and less manual work.
Here, it is not just configs — it's a structured #AI collaboration framework that streamlines development, from design to access control.
⬇️Discover more with #MindSporegitcode.com/mindspore/hype…
🚀 Train #AI models across devices and organizations more efficiently with ALI-DPFL.
Built on #MindSpore, it leverages adaptive local iterations to improve model performance — same privacy and communication constraints, better results.
⬇️Details here
arxiv.org/pdf/2308.10457
When it comes to offline models and on-device inference, don't miss #MindSpore Lite.
➡️ Unified model IR (MindSpore / ONNX / TF)
➡️ Feature-rich, efficient runtime
➡️ High-performance CPU operator libraries
Build and deploy #AI models with ease👇
mindspore.cn/tutorials/en/r…
⚡ Meet DVM — making dynamic compilation faster and lighter.
With bytecode instead of machine code, DVM describes computations at runtime for efficient execution, now on #MindSpore.
☑️ Higher performance
☑️ Lower overhead
Dive into the paper ↓
arxiv.org/abs/2603.24239
Tell your autoregressive tracker where to look and how to describe with ARTrackV2 👀
A unified generative framework for target localization and appearance modeling — less complexity, more efficiency.
Explore ARTrackV2 on #MindSpore 🚀
github.com/MIV-XJTU/ARTra…#MindSporeCaseStudy
What happens between pure noise and a generated image?
This #MindSpore Case Study walks through the core ideas behind #DDPM diffusion models — from theory to implementation👇mindspore.cn/tutorials/en/r…
#MindSpore 2.9 is bringing something innovative to automatic operator fusion ⚡
Meet Graph-Free Fusion ⬇️
Fewer constraints, higher performance, and easier optimization
Plus, to better support it, DVM is now fully open-sourced.
Dive in 👉 mindspore.cn/version-update…
Curious about #MindSpore 2.9?
What's new for developers:
⚡ Faster dynamic graph execution with Graph-Free Fusion
🧩 More flexible custom ops and execution control
🖥️ Better large-scale training stability
🧪 New MindSpore Science agent system
Get started:
mindspore.cn/en
👀 If you've been exploring quantum computing workflows...
Scaling quantum simulation, exploring richer algorithms, and connecting to real hardware —
#MindSpore Quantum 0.12.0 is here to make everything smoother. ⚛️
See what's new 👇
atomgit.com/mindspore/mind…#opensource
How can foundation models tackle multi-scale and multi-physics scientific computing?
OmniArch comes with:
⚡ 1D-2D-3D joint pretraining
⚡ A novel PDE-Aligner
⚡ Strong in-context and zero-shot capabilities
👇Try it on #MindSporeopeni.pcl.ac.cn/cty315/OmniArc…#MindSporeCaseStudy
🚀 A new chapter begins with #MindSpore 2.9
Not just an #AI framework upgrade —
we're making model training faster ⚡
operator development easier 🛠️
and scientific workflows smarter 🧠
👉 Start exploring here
mindspore.cn/install/en
📌 More feature deep dives coming soon!
Struggling with Host-side lag or CPU chaos? 😵💫
See how far #MindSpore has come with CPU affinity.
From scattered tricks → a complete runtime optimization system ⚙️
Built for real Host-bound workloads.
Start optimizing👇
mindspore.cn/tutorials/en/m…
More to come in MindSpore 2.9 ✨
From symmetric graphs to structural inclusion — the structure-inclusive similarity (SIS) improves GNN-based prediction of drug-target affinity. 💊
➡️ More accurate, higher performance
👇Discover more and try it on #MindSporeacademic.oup.com/bioinformatics…#MindSporeCaseStudy
29 Followers 418 FollowingStudent at @Tsinghua_uni
working to make dad proud and learning about the world.
"I can't control everything, but my self-belief and my self-love are up to me."
2K Followers 2K FollowingSenior AI reporter at South China Morning Post. RT not endorsement. DM for Signal, where conversations are always off the record until agreed otherwise.
189 Followers 3K Followingo vitae philosophia dux, o virtutis indagatrix expultrixque vitiorum! quid non modo nos, sed omnino vita hominum sine te esse potuisset? (Cic. Tusc. 5.5) #GIABO
130K Followers 267 FollowingThe AI Lab behind GLM models, dedicated to inspiring the development of AGI to benefit humanity.
https://t.co/7a5aSCUfnr
https://t.co/x14hb3jO7O
651 Followers 4K FollowingJust a guy from New Delhi,sharing this world with you. I hope my opinions & actions are a source of Good to you(&others)both in this life & hereafter.RT≠Endorse
3K Followers 281 FollowingFast AI inference. 200+ models via one API. OpenAI/Anthropic compatible.
💬 Join Community → https://t.co/5AzVHJpDHx
🚀 Get Higher Rate Limits → https://t.co/5EvDyDEbaC
9K Followers 358 FollowingOpenBMB (Open Lab for Big Model Base) aims to build foundation models and systems towards AGI.
Connect with us: https://t.co/N9pevTnoOa
8K Followers 1K FollowingWelcome to our Official OrangePi Twitter account. The World's Leading Open-Source Single-Board Computer Producer.
Join telegram https://t.co/NKxsmYEwxc…
2 Followers 0 FollowingMakerizon is an interdisciplinary team combining innovation and tech development, promoting industry-academia collaboration and driving local economic growth
111K Followers 523 FollowingSecure enterprise open source for the world.
We're also the publisher of @Ubuntu, the most popular and reliable Linux OS.
#AI #MLOps #Robotics #IoT #Cloud
191K Followers 616 FollowingGet all the latest news and updates for developers from Huawei. Turn your app ideas into reality.
Privacy Statement Link:https://t.co/8I4xWHpqSp
1K Followers 110 FollowingShan Shui Conservation Centre is a Chinese NGO founded in 2007. We practice community-based conservation and citizen science to conserve species and ecosystems.
1.5M Followers 276 FollowingThe engine room of @Google. Building AI safely and responsibly to solve the world’s most complex problems. Join us: https://t.co/jUHQA27iBL
829K Followers 344 FollowingTogether with the AI community, we are pushing the boundaries of what’s possible through open science to create a more connected world.
178K Followers 613 FollowingBienvenue sur le compte officiel de Huawei France ! Retrouvez ici toutes les actualités institutionnelles de #Huawei en France et dans le monde #Innovation
51 Followers 42 Following🇫🇷 HUAWEI Developer Groups (HDG) est une communauté pour les développeuses et développeurs intéressé.e.s et passionné.e.s par les nouvelles technologies.
505K Followers 87 FollowingTensors and neural networks in Python with strong hardware acceleration. PyTorch is an open source project at the Linux Foundation. #PyTorchFoundation
5.1M Followers 4 FollowingOpenAI’s mission is to ensure that artificial general intelligence benefits all of humanity. We’re hiring: https://t.co/dJGr6LgzPA
65K Followers 249 FollowingGlobal non-profit: promotes and protects Open Source software, development & communities through education and advocacy. Yes, we coined the term!