ModelArts Standard
The full one-stop platform: ExeML, data management, dev environments, training, model management and deployment.
For full AI development
Category · AI & Machine Learning
ModelArts AI platform
A one-stop AI development platform — build, train and deploy models faster, with DDS on Huawei Cloud, hosted in MENA.
+50%
Faster development
1T+
Model parameters
<0.5%
Training failure rate
30 days
Uninterrupted 10k-card runs
01 — Overview
One platform for the whole AI lifecycle — guided by DDS
ModelArts is Huawei Cloud's one-stop AI development platform. With large-volume data preprocessing, semi-automated data labeling, distributed training, automated model building, and on-demand deployment across devices, edge and cloud, it helps your team quickly build and deploy models and manage the entire AI development lifecycle.
It covers every stage — data processing, algorithm development, and model training and deployment — on a three-layer architecture (compute, AI platform, and development toolchain) that supports TensorFlow, PyTorch, MindSpore and custom frameworks. DDS sets it up on Huawei Cloud, sized for your workload and hosted close to your users in MENA, and supports your team end to end.
02 — The ModelArts family
Pick the right entry point
ModelArts is delivered as several products — from the full platform to compute-only and edge options — so you start where it fits.
ModelArts Studio (MaaS)
End-to-end toolchain for foundation-model production, plus compute and popular open-source models.
For production-ready foundation models
ModelArts Lite Server
Cost-effective AI compute on bare-metal servers via EIP, with mainstream AI kits and Huawei acceleration plugins.
For teams that need only compute
ModelArts Lite Cluster
Direct access to native Kubernetes APIs to manage nodes and clusters within resource pools.
For Kubernetes-savvy teams
ModelArts Edge
Simplifies edge deployment: model and node management, resource pooling, load balancing and commercial-use assurance.
For edge + cloud scenarios
AI Gallery
A high-quality model development experience with access to a wealth of community resources.
For AI development & exploration
03 — Benefits & business value
What ModelArts does for your business
End-to-end tool chain
DataOps, MLOps and DevOps in one platform boost development efficiency by up to 50%.
Business impactYour team ships AI features faster, turning ideas into production sooner.
Massive-scale training
Run ultra-large distributed jobs — models with up to a trillion parameters over petabytes of data.
Business impactYou tackle the most ambitious models without hitting an infrastructure ceiling.
High reliability
Jobs auto-recover from faults with a sub-0.5% failure rate; 10,000-card clusters run uninterrupted for 30 days.
Business impactLong training runs finish on time, protecting timelines and compute spend.
Simplified development (AutoML)
Zero-code ExeML automates model design, tuning, compression and deployment.
Business impactDomain experts build models without deep ML expertise, widening who can deliver value.
Flexible deployment
Deploy to cloud or edge — real-time, batch or edge inference with elastic auto-scaling.
Business impactYou serve models where the business needs them, at the right cost and latency.
Pay as you go
Pay-per-use, or a discounted yearly/monthly subscription — price depends on the compute and configuration you choose.
Pay-per-use · on demandYearly / monthly · discounted
Business impactYou match AI spend to real usage and avoid large upfront GPU/NPU investment.
04 — Use cases
Built for real AI workloads
Foundation models
Intelligent Q&A, chatbots, automated summarization, machine translation and text classification.
AIGC
AI-assisted generation of images, copy, audio and video.
Autonomous driving
Environment perception, path planning and driving control.
Finance & government
Efficient, accurate services for financial institutions and AI-powered public-service decisions.
Mining & railway
Safer, smarter production; intelligent train scheduling, fault prediction and rail security.
Healthcare & more
Automated report interpretation, internet diagnoses and health management — across finance, government, mining, railway, healthcare and automotive.
05 — FAQ
ModelArts, answered
What is ModelArts in simple terms?
It's a one-stop cloud platform for building AI — covering data, training and deployment in one place, so you don't manage the infrastructure yourself.
Do I need to be an AI expert to use it?
No. Zero-code AutoML (ExeML) lets non-experts build models, while notebooks and frameworks (PyTorch, TensorFlow, MindSpore) support advanced teams.
Which frameworks does it support?
Mainstream frameworks including PyTorch, TensorFlow and MindSpore, with JupyterLab or remote-IDE development.
How are models deployed?
As real-time, batch or edge inference, with elastic auto-scaling so services stay fast and cost-efficient under changing load.
How big can training jobs get?
Ultra-large scale — models up to a trillion parameters across petabytes of data, with one-click distributed training.
How does DDS help?
DDS sets up ModelArts on Huawei Cloud in MENA, sizes the compute for your workload, and supports your team across the AI lifecycle.
Build AI with DDS in MENA
Talk to a DDS AI architect about platform setup, compute sizing and getting your models to production.
Free AI workload assessment
EGP / SAR pricing & local billing
Reply within one business day