Deploy stdapi.ai on AWS¶

Get a production-grade OpenAI-compatible AI gateway running on AWS in 5 minutes. Terraform handles everything — ECS Fargate, HTTPS, auto-scaling, optional WAF, and optional monitoring.

Quick Start¶

Prerequisites¶

1. Subscribe to stdapi.ai on AWS Marketplace (14-day free trial included).
2. Install Terraform or OpenTofu >= 1.5.
3. Configure AWS credentials (aws configure or aws sso login).

aws sts get-caller-identity
aws configure get region

Deploy¶

git clone https://github.com/stdapi-ai/samples.git
cd samples/getting_started_production/terraform
terraform init
terraform apply

curl -L https://github.com/stdapi-ai/samples/archive/refs/heads/main.zip -o samples.zip
unzip samples.zip
cd samples-main/getting_started_production/terraform
terraform init
terraform apply

That's it. In ~5 minutes you have:

• Production-grade ECS Fargate deployment with HTTPS
• Optional WAF protection with rate limiting
• Optional CloudWatch alarms and monitoring
• Auto-scaling and API key authentication
• Interactive API documentation at /docs
• IP-restricted access (your IP only)

%%{init: {'flowchart': {'htmlLabels': true}} }%%
flowchart LR
  openai["<img src='../styles/logo_openai.svg' style='height:64px;width:auto;vertical-align:middle;' /> OpenAI SDK"] -->|HTTPS| alb["<img src='../styles/logo_amazon_load_balancing.svg' style='height:64px;width:auto;vertical-align:middle;' /> ALB"]
  anthropic["<img src='../styles/logo_anthropic.svg' style='height:64px;width:auto;vertical-align:middle;' /> Anthropic SDK"] -->|HTTPS| alb
  alb --> ecs["<img src='../styles/logo.svg' style='height:64px;width:auto;vertical-align:middle;' /> stdapi.ai<br/>ECS Fargate"]
  ecs --> bedrock["<img src='../styles/logo_amazon_bedrock.svg' style='height:64px;width:auto;vertical-align:middle;' /> AWS Bedrock"]
  ecs --> polly["<img src='../styles/logo_amazon_polly.svg' style='height:64px;width:auto;vertical-align:middle;' /> AWS Polly"]
  ecs --> transcribe["<img src='../styles/logo_amazon_transcribe.svg' style='height:64px;width:auto;vertical-align:middle;' /> AWS Transcribe"]
  ecs --> s3["<img src='../styles/logo_amazon_s3.svg' style='height:64px;width:auto;vertical-align:middle;' /> AWS S3"]
  ecs --> cloudwatch["<img src='../styles/logo_amazon_cloudwatch.svg' style='height:64px;width:auto;vertical-align:middle;' /> CloudWatch"]

Get Your Credentials¶

terraform output -raw api_key
terraform output api_endpoint
terraform output docs_url

Make Your First API Call¶

Want to explore the API without writing code? Use the docs_url from terraform output docs_url to open the interactive Swagger documentation in your browser — you can browse all available endpoints and make live API calls directly from the page, no code required.

stdapi.ai is compatible with both OpenAI and Anthropic SDKs. If you've used either before, you already know how to use it — only the base URL changes. Here are the raw HTTP calls with curl so you can verify the endpoint from any shell:

API_ENDPOINT=$(terraform output -raw api_endpoint)
API_KEY=$(terraform output -raw api_key)

curl "$API_ENDPOINT/v1/chat/completions" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "amazon.nova-micro-v1:0",
    "messages": [{"role": "user", "content": "Hello! Tell me a joke."}]
  }'

API_ENDPOINT=$(terraform output -raw api_endpoint)
API_KEY=$(terraform output -raw api_key)

curl "$API_ENDPOINT/anthropic/v1/messages" \
  -H "x-api-key: $API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "amazon.nova-micro-v1:0",
    "max_tokens": 1000,
    "messages": [{"role": "user", "content": "Hello! Tell me a joke."}]
  }'

Using the official SDKs? Point the base_url / base_url option at $API_ENDPOINT/v1 (OpenAI SDK) or $API_ENDPOINT/anthropic (Anthropic SDK) and use your existing code — no other changes needed. The API Overview has SDK snippets for Python, Node.js, and more.

No API key configured? stdapi.ai runs without authentication by default for quick testing. Add api_key_create = true to your Terraform config to enable it.

Available models: All AWS Bedrock models available in your configured regions are automatically discovered and exposed. List them with GET /search_models — the default endpoint for model discovery, returning full details (capabilities, routes, regions) for every model. Use query parameters to filter by capability (modality, route, region, streaming, legacy status) — particularly useful for AI agents that need to pick the right model dynamically. GET /v1/models is also available for strict OpenAI SDK compatibility.

Verify the deployment is healthy:

curl https://your-endpoint/health
# → {"status": "ok"}

The /health endpoint requires no authentication and is used by the ALB health check.

Troubleshooting¶

The two most common first-deployment hiccups:

Full troubleshooting guide: Troubleshooting — 403 auth errors, 404 model not found, ThrottlingException, S3 errors, VPC connectivity, Terraform IAM failures, and more.

Next Steps¶