Overview
AI Suite AI Server turns your Amazon EKS cluster into a private AI platform. Run OpenAI-compatible APIs for chat, embeddings, image generation, speech synthesis, transcription, and vision inside infrastructure you control. The Helm delivery installs AI Gateway and AI Server workers with health probes, graceful draining, persistent model storage, NetworkPolicy defaults, and production-ready Kubernetes resources.
Keep prompts, documents, model outputs, and API traffic within your AWS environment. Connect AI Suite apps, OpenAI-compatible tools, and custom applications through one authenticated API; scale GPU workers independently while AI Gateway routes by health, load, latency, and model availability; monitor nodes with the built-in dashboard and Prometheus metrics; and roll workers without dropping in-flight requests.
This is a bring-your-own-license (BYOL) offer. A Software Tailor commercial AI Server license is required for each active worker node. AWS separately bills EKS compute, storage, networking, and GPU capacity. Customers are responsible for TLS ingress, API keys, capacity planning, and AWS infrastructure charges. The software contacts registration.softwaretailor.com for license activation unless configured for air-gapped operation.
Highlights
- Run private OpenAI-compatible chat, embedding, image, speech, transcription, and vision APIs on Amazon EKS.
- Scale AI Server workers behind AI Gateway with health-aware routing, graceful draining, and Kubernetes-native operations.
- Keep workloads in your AWS environment with BYOL licensing, authenticated APIs, NetworkPolicy defaults, and Prometheus monitoring.
Details
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
Pricing
Vendor refund policy
No refund
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
AI Suite AI Server for Amazon EKS
- Amazon EKS
Helm chart
Helm charts are Kubernetes YAML manifests combined into a single package that can be installed on Kubernetes clusters. The containerized application is deployed on a cluster by running a single Helm install command to install the seller-provided Helm chart.
Version release notes
Initial AWS Marketplace release of AI Suite AI Server and AI Gateway for Amazon EKS. Includes OpenAI-compatible APIs, BYOL licensing, health-aware gateway routing, graceful draining, persistent worker storage, NetworkPolicy defaults, and a stub engine for installation validation.
Additional details
Usage instructions
Deploy AI Suite AI Server 2.5.0 to an existing Amazon EKS cluster with Helm 3.19 or later.
Prerequisites:
- The AWS account has subscribed to this product, and the AWS CLI is configured for that account.
- kubectl is configured for the target EKS cluster.
- The EKS node IAM role can pull subscribed images from Amazon ECR.
- You have an AI Suite Commercial license GUID.
- You have generated separate gateway-client and gateway-to-worker key stores as described at https://softwaretailor.com/docs/ai-server/index.htm .
Create the namespace and protected Kubernetes Secrets. Replace the paths and placeholder value; do not commit the license or raw keys to source control.
kubectl create namespace aisuite kubectl -n aisuite create secret generic aisuite-license \ --from-literal=key='<AI_SUITE_LICENSE_GUID>' kubectl -n aisuite create secret generic aisuite-gateway-client-keys \ --from-file=server-keys.json=./gateway/credentials/server-keys.json kubectl -n aisuite create secret generic aisuite-worker-keys \ --from-file=server-keys.json=./worker/credentials/server-keys.jsonCreate a local aisuite-values.yaml. The worker token is the raw value associated with the key in the worker key store; it is not the hashed value stored in server-keys.json.
license: existingSecret: aisuite-license gateway: existingClientKeysSecret: aisuite-gateway-client-keys pool: dnsDiscovery: enabled: true workerToken: '<RAW_GATEWAY_TO_WORKER_KEY>' worker: existingKeysSecret: aisuite-worker-keys engine: stubAuthenticate Helm to AWS Marketplace ECR, then pull and install the subscribed chart. The ECR authorization token is valid for 12 hours.
aws ecr get-login-password --region us-east-1 \ | helm registry login --username AWS --password-stdin \ 709825985650.dkr.ecr.us-east-1.amazonaws.com helm pull oci://709825985650.dkr.ecr.us-east-1.amazonaws.com/software-tailor/aisuite-helm \ --version 0.1.4 helm install aisuite ./aisuite-helm-0.1.4.tgz \ --namespace aisuite -f ./aisuite-values.yamlWait for the deployment, then verify health through a temporary local port forward:
kubectl -n aisuite rollout status deployment/aisuite-aisuite-gateway kubectl -n aisuite rollout status statefulset/aisuite-aisuite-worker kubectl -n aisuite port-forward service/aisuite-aisuite-gateway 8080:8080 curl http://127.0.0.1:8080/livez curl http://127.0.0.1:8080/readyzThe default stub engine validates the installation without a GPU or model download. For real inference, change worker.engine, storage, node selection, and GPU settings in the values file. The gateway is a private ClusterIP; add a TLS-terminating ingress or private load balancer before allowing client traffic. The software contacts registration.softwaretailor.com for BYOL seat activation unless license.registrationUrl is set to an empty value for an air-gapped deployment.
The chart's packaged values.yaml documents every configurable parameter. Common settings include gateway and worker replicas/resources, persistence, node selection, GPU scheduling, routing, autoscaling, probes, and NetworkPolicy. AWS Marketplace regionalizes the two image repositories and tags from gateway.image and worker.image; keep those Marketplace-provided image values so every container is pulled from the subscribed product rather than an external registry.
Resources
Vendor resources
Support
Vendor support
Documentation and deployment guidance are available at https://softwaretailor.com/docs/ai-server/index.htm . For product, licensing, and technical support, email support@softwaretailor.com . Include your AI Suite AI Server version, deployment environment, and a concise problem description. Do not include license keys, API keys, passwords, or other secrets in support messages.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products
