AI tools & infrastructure on Alibaba Cloud
AI applications, model operations, inference and supporting vector databases.
4 documented options on Alibaba Cloud. Provider coverage reflects reviewed evidence, not an exhaustive compatibility claim.
Read this page as JSON or Markdown.
| Listing | Documented clouds | Deployment and infrastructure | Licensing | Installation | Operations and limitations |
|---|---|---|---|---|---|
| Dify Self-Hosted | Alibaba Cloud | Customer-operated Servers / containers | Source Available Modified Apache 2.0 terms; multi-workspace service operation and frontend branding can require commercial authorization. | Official Docker Compose deployment; pin a release and configure its required services. | Customer operates the containers, persistent stores, secrets, backups and upgrades; commercial edition support varies.Scope and evidenceSelf-hosting Dify does not itself host every model used by a flow. Model-provider plugins, embedding services, tools and connectors may send inputs or retrieved content outside the deployment. Select local endpoints where required and review each dependency. This entry does not conflate the generic self-hosted stack with the separately licensed AWS Premium edition. Dify 1.20.0 changes deployment and API behavior: OPENAPI_ENABLED defaults to true, shared.env is retired, six migrations apply from 1.19, and HITL, plugin-daemon, OpenShell, and optional file-scanning requirements change. It also fixes OAuth callback reuse, member-role enforcement, vector deletion scope, public stream detail exposure, and CVE-2026-116591.
Cloud evidence reviewed 2026-11-12 |
| MaxKB Community Edition | Alibaba Cloud | Customer-operated Servers / containers | Open Source / Open Core GNU General Public License version 3 for Community Edition. Professional and Enterprise editions add separately licensed features and support. | Create the documented MaxKB Community Edition service instance through Alibaba Cloud Compute Nest, or use the publisher-documented Alibaba Cloud Marketplace image on a customer ECS server. | Customer manages the ECS host, VPC and security-group exposure, administrator credentials, persistent PostgreSQL data, backups, upgrades, capacity and credentials for model and tool integrations.Scope and evidenceThe Compute Nest deployment requires RAM permissions with full access to ECS, VPC and ROS plus Compute Nest user access; review and reduce deployment privileges after provisioning where practical. Community Edition is single-tenant, x86-64 only and supports single-node or cold-standby deployment rather than Enterprise clustering. RBAC, SSO, audit logs, open API access, conversation-user authorization and several enterprise integrations are reserved for paid editions. Current upstream v2 installation guidance calls for at least 4 CPU cores, 8 GB RAM and 100 GB of disk, exposes the web service on port 8080 and supplies default administrator credentials that must be changed immediately. The publisher recommends offline installation for production and requires a database backup before upgrades. Public or separately hosted model, embedding, reranking and tool providers can receive application data even when MaxKB itself runs in the customer account. The current upstream v2 container image and persistence paths differ from the legacy English guide; check the release packaged by the cloud installer before applying upstream configuration or upgrades. MaxKB v2.10.9-lts fixes multiple cross-workspace access and deletion flaws, stored XSS with account-takeover impact, custom-tool secret exposure, SSRF paths including possible cloud-credential theft, anonymous provider API-key disclosure and unauthorized conversation or knowledge-base access. Upgrade the LTS line and retest workspace, tool and provider boundaries.
Cloud evidence reviewed 2026-11-16 |
| NVIDIA AI Enterprise | AWS, Azure, Google Cloud, Oracle Cloud, Alibaba Cloud, Tencent Cloud, Volcano Engine | Customer-operated Servers / containers | Commercial Only Commercial NVIDIA AI Enterprise subscription, consumption or perpetual license; most customer-cloud paths require BYOL. | NVIDIA AI Enterprise or GPU-optimized cloud VM images, standard instances with BYOL, or managed Kubernetes on supported cloud providers | Customer manages cloud GPU instances or Kubernetes, drivers, GPU Operator and container runtime, networking, storage, identities, NGC access, model services, monitoring, upgrades and license activation.Scope and evidenceThis entry covers customer-operated NVIDIA AI Enterprise software, not a managed AI service. A license is required and is not included with most standard, GPU-optimized VMI or managed-Kubernetes deployments; on-demand NVIDIA AI Enterprise VMIs and private offers use separate included-license processes. Not every component is available on every cloud or deployment model. GPU deployments are generally licensed per GPU, while CPU-only instances require an instance license. Tencent Cloud and ByteDance Volcano Engine have documented standard-instance BYOL paths. These guides require certified software and supported instance types; they do not establish that every GPU, component or provider image is eligible. Their per-instance wording differs from the general per-GPU licensing guidance, so confirm the applicable licensing unit. The November lifecycle matrix lists active 9.0/R620, 8.3/R595, and 7.9 LTSB/R580 branches with different end dates. NeMo Microservices support ended October 1, 2026, after 25.12 became the final GA release; select and patch each component branch independently.
Cloud evidence reviewed 2026-11-16 |
| RAGFlow | AWS, Azure, Google Cloud, Alibaba Cloud, Railway | Customer-operated Servers / containers | Open Source / Open Core Apache License 2.0 for upstream RAGFlow; models, data assets and dependencies retain their own terms. | Versioned upstream Docker Compose, provider-specific VM packages or community deployment templates; select a path compatible with the intended RAGFlow release. | Customer manages document storage, indexing, app services, access controls, model endpoints, backups and upgrades.Scope and evidenceDeployment options include customer AWS, Azure and Google Cloud infrastructure, Alibaba Compute Nest, and a Railway project. The evidence trail identifies the publishers and template versions. Hosting RAGFlow does not automatically self-host its LLM, embeddings or reranker; choose private or local model endpoints where required.
Cloud evidence reviewed 2026-11-12 |
Deployment evidence and category fit
Dify Self-Hosted
Visual AI application and RAG workflow platform available as a self-hosted container stack.
Self-hosting Dify does not itself host every model used by a flow. Model-provider plugins, embedding services, tools and connectors may send inputs or retrieved content outside the deployment. Select local endpoints where required and review each dependency. This entry does not conflate the generic self-hosted stack with the separately licensed AWS Premium edition. Dify 1.20.0 changes deployment and API behavior: OPENAPI_ENABLED defaults to true, shared.env is retired, six migrations apply from 1.19, and HITL, plugin-daemon, OpenShell, and optional file-scanning requirements change. It also fixes OAuth callback reuse, member-role enforcement, vector deletion scope, public stream detail exposure, and CVE-2026-116591.
- Dify Docker deployment on Alibaba Cloud ECS ↗
Provider guide · Reviewed 2026-11-12. Vendor support commitment: not assessed; confirm the selected release and contract.
Alibaba Cloud documents cloning upstream Dify and running its Docker deployment on a customer ECS instance. The example then integrates an MCP server; model providers and other outbound integrations must be reviewed separately.
- Dify Community Edition full application on Alibaba Compute Nest ↗
Provider guide · Reviewed 2026-11-12. Vendor support commitment: not assessed; confirm the selected release and contract.
Alibaba documents a Dify Community Edition service instance that creates ECS compute in the customer-selected VPC, zone and region, then exposes the installed Dify application for administrator registration. This is application hosting rather than an MCP/model connector, and is separate from the licensed Dify AWS Premium offering. Review model endpoints and release compatibility.
Cloud evidence reviewed 2026-11-12. Full listing and qualifications
MaxKB Community Edition
Customer-hosted agent and workflow platform with knowledge bases, MCP and tool integrations, and configurable language, embedding, reranking and multimodal models.
The Compute Nest deployment requires RAM permissions with full access to ECS, VPC and ROS plus Compute Nest user access; review and reduce deployment privileges after provisioning where practical. Community Edition is single-tenant, x86-64 only and supports single-node or cold-standby deployment rather than Enterprise clustering. RBAC, SSO, audit logs, open API access, conversation-user authorization and several enterprise integrations are reserved for paid editions. Current upstream v2 installation guidance calls for at least 4 CPU cores, 8 GB RAM and 100 GB of disk, exposes the web service on port 8080 and supplies default administrator credentials that must be changed immediately. The publisher recommends offline installation for production and requires a database backup before upgrades. Public or separately hosted model, embedding, reranking and tool providers can receive application data even when MaxKB itself runs in the customer account. The current upstream v2 container image and persistence paths differ from the legacy English guide; check the release packaged by the cloud installer before applying upstream configuration or upgrades. MaxKB v2.10.9-lts fixes multiple cross-workspace access and deletion flaws, stored XSS with account-takeover impact, custom-tool secret exposure, SSRF paths including possible cloud-credential theft, anonymous provider API-key disclosure and unauthorized conversation or knowledge-base access. Upgrade the LTS line and retest workspace, tool and provider boundaries.
- MaxKB Community Edition service instance in Alibaba Compute Nest ↗
Provider guide · Reviewed 2026-11-16. Vendor support commitment: not assessed; confirm the selected release and contract.
Alibaba Cloud documents creation of a MaxKB Community Edition service instance in a customer-selected region, zone, VPC and VSwitch using customer-selected ECS capacity and billing. The deployed application is reached through the ECS public address on port 8080. The workflow requires broad ECS, VPC, ROS and Compute Nest permissions and returns default administrator credentials that must be changed.
- MaxKB image deployment from Alibaba Cloud Marketplace ↗
Vendor / upstream documentation · Reviewed 2026-11-16. Vendor support commitment: not assessed; confirm the selected release and contract.
The publisher documents selecting the MaxKB image while purchasing an Alibaba Cloud server, starting the image and opening port 8080 in the customer's security group. It recommends at least 4 CPU cores and 8 GB RAM and instructs administrators to replace the default password after first login.
Cloud evidence reviewed 2026-11-16. Full listing and qualifications
NVIDIA AI Enterprise
Provides an enterprise-supported suite of AI frameworks, infrastructure software, operations components and development tools.
This entry covers customer-operated NVIDIA AI Enterprise software, not a managed AI service. A license is required and is not included with most standard, GPU-optimized VMI or managed-Kubernetes deployments; on-demand NVIDIA AI Enterprise VMIs and private offers use separate included-license processes. Not every component is available on every cloud or deployment model. GPU deployments are generally licensed per GPU, while CPU-only instances require an instance license. Tencent Cloud and ByteDance Volcano Engine have documented standard-instance BYOL paths. These guides require certified software and supported instance types; they do not establish that every GPU, component or provider image is eligible. Their per-instance wording differs from the general per-GPU licensing guidance, so confirm the applicable licensing unit. The November lifecycle matrix lists active 9.0/R620, 8.3/R595, and 7.9 LTSB/R580 branches with different end dates. NeMo Microservices support ended October 1, 2026, after 25.12 became the final GA release; select and patch each component branch independently.
- NVIDIA AI Enterprise on Oracle Cloud and Alibaba Cloud ↗
Vendor / upstream documentation · Reviewed 2026-11-16. Vendor support commitment: not assessed; confirm the selected release and contract.
NVIDIA's current cloud matrix includes standard instances on Oracle Cloud Infrastructure and Alibaba Cloud, NVIDIA AI Enterprise VM images on OCI and GPU-optimized VM images on Alibaba Cloud. Customers must confirm the supported component set, certified instance and license path.
Cloud evidence reviewed 2026-11-16. Full listing and qualifications
RAGFlow
Self-hosted document ingestion, retrieval and context infrastructure for AI applications.
Deployment options include customer AWS, Azure and Google Cloud infrastructure, Alibaba Compute Nest, and a Railway project. The evidence trail identifies the publishers and template versions. Hosting RAGFlow does not automatically self-host its LLM, embeddings or reranker; choose private or local model endpoints where required.
- RAGFlow Community deployment on Alibaba Cloud Compute Nest ↗
Provider guide · Reviewed 2026-11-12 · Scope: Community / upstream open-source; third-party distribution where noted. Vendor support commitment: not assessed; confirm the selected release and contract.
Alibaba Cloud documents provisioning a RAGFlow Community service instance with customer ECS, VPC and vSwitch parameters. This is distinct from the managed AI-Native Database Service integration. Review image version, access controls, model endpoints and the broad example RAM permissions before production.
Cloud evidence reviewed 2026-11-12. Full listing and qualifications
Deployment and procurement checks
- Dedicated connectivity
- Express Connect provides a dedicated on-premises-to-VPC connection. Confirm the access point, routing and application region separately from the deployment package. Provider source ↗ · Reviewed 2026-11-14
For every provider, confirm marketplace/private-offer terms, commitment eligibility, regional service availability, private connectivity and any vendor access against your selected deployment. Provider capabilities do not automatically establish application support.
Plan this deployment
Account boundary: Alibaba Cloud account and resource group. Access: Resource Access Management (RAM). Network: VPC.
Provider documentation ↗ · Guidance reviewed 2026-11-10. Confirm the application edition, region, support access, backup locations and operating responsibilities before choosing.
Compare this category across clouds · Search these listings · Compare 2–4 listings