Why Alibaba Cloud
Alibaba Cloud’s full-stack solution for generative AI (GenAI) provides whole-process services for foundation models (FMs) and other AI development tasks. This solution helps you build and optimize FMs, fine-tune them according to your business preferences, and deploy them easily as online services, all on purpose-built AI infrastructure optimized for performance and efficiency. Regardless of the scale and stage of your business, this solution enables you to create new and intelligent customer experiences and drive business transformation with innovations in generative AI.
-
Ready-to-Use AI Computing Services
Kickstart large-scale FM training and generative AI application development with our purpose-built AI services featuring high performance, scalability, and cost-effectiveness
-
A Diverse Selection of Open-Source FMs
Choose FMs for your business from industry-leading open-source models including Alibaba Cloud's Tongyi Qianwen (Qwen), Stable Diffusion, Llama 2, and more from mainstream AI communities
-
High-Performance FM Training and Inference
Apply end-to-end performance optimization to accelerate dataset processing, as well as model training and inference, and enhance it with a GPU-based AI acceleration solution
-
Business Transformation at the Forefront of AI
Build innovative applications with the power of generative AI to transform customer experience, boost productivity, and stay at the forefront of industry development, like Futureverse did
Disclaimer
Some images displayed in this page are solely for the purpose of demonstrating product functions. All information shown in these images has been generated by AI models and is not verified by Alibaba Cloud. Alibaba Cloud makes no warranties, expressed or implied, as to the authenticity, accuracy and completeness of all such information.
2023 Alibaba Cloud Global Summit: Powering Up Future Business with Generative AI
At the 2023 Alibaba Cloud Global Summit, Mr. Dongliang Guo, VP of Product and Solution, discussed the significance of the cloud computing industry when driving AI transformation, introduced Alibaba Cloud’s latest innovations (including Cloud-Native AI suite 2.0 and the 8th Generation of ECS), and demonstrated the exciting capabilities of multi-modality content generation jointly supported by Alibaba Cloud and key partners.
Learn MoreLearn More About Generative AI on Alibaba Cloud
A Wide Selection of Open-Source FMs
You can train and fine-tune popular FMs including Alibaba Cloud’s Qwen series in Platform for AI (PAI) in a few clicks, and easily deploy them as online services with PAI-EAS.
Tongyi Qianwen (Qwen)
Alibaba Cloud provides a series of open-source Tongyi Qianwen models: Qwen, the LLM; Qwen-VL, the large vision and language model; and Qwen-Audio, the large audio language model. Qwen models are pre-trained on multilingual data covering various industries and domains, and offer a wide range of capabilities, including multimodal understanding and generation, state-of-the-art image processing, and fully managed APIs to support your innovation in generative AI.
You can easily fine-tune Qwen models with your enterprise data and deploy them as online services that understand your business.
Llama 2
Llama 2 is an open-source LLM with a larger set of training data and a longer context length than its predecessor, Llama 1, making it more versatile and accurate in generating natural language at scale. You can access, fine-tune, and deploy Llama 2 with Platform for AI (PAI) in a few simple steps.
Stable Diffusion
Stable Diffusion is an open-source model that leverages advanced deep-learning algorithms and techniques to generate visually compelling images based on text descriptions. You can access, fine-tune, and deploy Stable Diffusion with Platform for AI (PAI) in a few simple steps.
ChatGLM
ChatGLM is an open-source LLM that uses supervised fine-tuning (SFT), a self-feedback system, and human feedback reinforcement learning (RLHF) to better align with human preferences. You can access, fine-tune, and deploy Stable Diffusion with Platform for AI (PAI) in a few simple steps.
Platform for AI (PAI): Whole-Process AI Engineering
PAI provides an end-to-end optimization solution to streamline AI engineering including PAI-iTAG for data labeling, PAI-DSW for model building, PAI-DLC for model training, and PAI-EAS for model inference and deployment.
Learn More About PAI
Data Preparation
Ready your data for model training with intelligent, customizable, and highly efficient multimodal data labeling services
Model Development
Build foundation models with our one-stop visualized modeling tool - PAI-Designer, or perform interactive development with Notebook in PAI-DSW
Model Training
Train models with PAI-DLC, our one-stop platform for cloud-native deep learning and training compatible with predefined and customized algorithm frameworks.
Model Deployment
Deploy your model as an online service or a web app with PAI-EAS, which supports push-button deployment of large-scale complex models
Data Preparation
Ready your data for model training with intelligent, customizable, and highly efficient multimodal data labeling services
Model Development
Build foundation models with our one-stop visualized modeling tool - PAI-Designer, or perform interactive development with Notebook in PAI-DSW
Model Training
Train models with PAI-DLC, our one-stop platform for cloud-native deep learning and training compatible with predefined and customized algorithm frameworks.
Model Deployment
Deploy your model as an online service or a web app with PAI-EAS, which supports push-button deployment of large-scale complex models
AI Infrastructure Built for Performance
You can perform compute-intensive AI tasks such as FM training and customization on our performance-optimized AI infrastructure, or choose our Cloud-Native AI suite to improve the speed and efficiency of your AI workloads on Kubernetes.
Learn More About Alibaba Cloud's AI Services
Cloud-Native AI: A Kubernetes-Based Service That Accelerates AI Development
Cloud-Native AI helps you utilize cloud-native architectures and technologies to quickly develop AI-based systems in Alibaba Cloud Container Service for Kubernetes (ACK). It provides a set of essential features and services to help you accelerate AI workloads and simplify MLOps.
PAI-Lingjun Intelligent Computing Service: All-in-One Platform for Powerful AI Computing (Currently available only in Ulanqab, China)
PAI-Lingjun Intelligent Computing Service provides high-intensity heterogeneous computing services and AI capabilities for large-scale generative AI tasks based on the integrated optimization technology of software and hardware. It also leverages pre-packaged GPUs and Platform for AI (PAI) to provide full-link performance acceleration for AI training and inference tasks in one stop.
- High-performance Remote Direct Memory Access (RDMA) networks greatly accelerate AI training.
- Cloud Paralleled File System (CPFS) provides efficient and reliable storage services for AI training.
- Our distributed training acceleration engine accelerates dataset management, cloud computing, algorithms, scheduling, and cloud resources.
Unified Hardware and Software Acceleration for AI
Alibaba Cloud speeds up generative AI development with purpose-selected GPUs, GPU-oriented optimization and acceleration technologies for high-efficiency hardware utilization, and Platform for AI (PAI) optimization features for dataset processing, model training, and model inference tasks.
1. GPUs for Model Training and Inference
-
Model Training: gn7 series of ECS instances power large-scale training tasks with high-performance NVIDIA GPUs
Model Inference: gn6 series of ECS instances provide a cost-effective choice for model inference tasks
Learn More >
2. AI Acceleration
-
You can leverage GPU Accelerator AIACC to speed up AI training tasks by up to 70% and inference tasks by 2-3 times according to Stanford DAWN Deep Learning Benchmark.
Learn More >
Learn More About Generative AI on Alibaba Cloud
Retrieval-Augmented Generation (RAG)
Retrieval Augmentation Generation (RAG) is an architecture that augments the capabilities of an LLM like Qwen by adding an information retrieval system that provides the models with relevant contextual data. Alibaba Cloud’s vector data databases store, manage, and retrieve vector embedding data as high-dimensional data effectively and efficiently, making them the ideal choice for the information retrieval system of a RAG architecture.
Download Whitepaper
AnalyticDB for PostgreSQL
AnalyticDB for PostgreSQL is a vector database that enables gen-AI applications by providing efficient and effective storage, indexing, and querying of high-dimensional data. It also supports hybrid queries that combine scalar and vector data types.
Alibaba Cloud Elasticsearch
Alibaba Cloud Elasticsearch provides the k-nearest neighbor (kNN) search feature to meet your business requirements of vector search in different search scenarios, such as image searching, video fingerprinting, speech recognition, and commodity recommendation.
Hologres
Hologres is an all-in-one real-time data warehouse engine integrated with Platform for AI (PAI). The built-in vector search engine Proxima from DAMO Academy supports real-time feature data computing and vector retrieval.
OpenSearch
OpenSearch provides a high-performance, low-cost vector search solution, supporting multiple vector algorithms and real-time data updates and queries. This helps enterprises quickly build multimodal search services with LLM.
Innovative Applications Based on Generative AI
You can build generative AI applications to boost productivity, enable innovative content creation, or fulfill your other business needs, without concerns about the underlying architecture, like Tongyi Wanxiang.
Learn More About Building GenAI Applications on Alibaba Cloud
Prompt-Based Text Generation
Prompt-Based Image Generation
Similar Image Generation
Image Style Transfer
Clothes Virtual Try-On
Sketch-Based Image Generation
Tongyi Wangxiang Generating Images of Clothes Virtual Try-On Based on Image Input
Tongyi Wangxiang Generating Images of Clothes Virtual Try-On Based on Image Input
Learn More About Generative AI on Alibaba Cloud
Customer Success Stories
"We are excited to be working with the team at Alibaba Cloud. JEN is at the forefront of AI generative music and having access to their advanced on-demand training infrastructure and global partner network enables us to move fast and stay ahead in this rapidly evolving market."
Aaron McDonald, Founder and CEO, Futureverse
Futureverse is a New Zealand-based AI technology unicorn. Futureverse is working on JEN-1, their text-to-music generator that needs robust and reliable AI training infrastructure for the foundation model training. Alibaba Cloud's AI infrastructure achieved a high-performance score of 96% from a series of benchmark Proof of Concepts (PoCs) involving multiple cloud vendors. Alibaba Cloud also offers streamlined operational management based on integrated and standardized services for computing, storage, and networking, as well as full lifecycle capabilities of AI engineering, making it an excellent choice for foundation model training tasks like JEN-1.
Alibaba Cloud Keeps Driving Innovation in Generative AI
