The OpenAI-compatible API for AWS AI
Access 80+ Bedrock models (Claude, Llama, Nova) and AWS AI services (Polly, Transcribe, Translate) with your existing OpenAI clients—no code changes required.
stdapi.ai provides AWS AI infrastructure through an OpenAI-compatible interface. Connect your existing applications to over 40 Amazon Bedrock models and AWS AI services without changing code.
- 🔌 OpenAI Compatible – Applications designed for OpenAI's API work with stdapi.ai. Just change the endpoint.
- 🔒 Your Data, Your AWS Account – All conversations and data remain within your AWS environment. For HIPAA, GDPR, and regulatory compliance.
- 🌍 Multi-Region Architecture – Access models across multiple AWS Bedrock regions through one unified endpoint.
- 💰 Cost Optimized – Pay standard AWS rates with zero markup.
- ⚡ Enterprise Ready – CloudWatch integration, native guardrails, hardened container images, and production-ready infrastructure.
Access models from:
- Anthropic (Claude for reasoning and coding)
- Amazon (Nova family for cost-effective performance)
- OpenAI, Meta (Llama), Mistral AI, Google, Cohere, Stability AI, DeepSeek, Qwen, and more
Beyond Bedrock language models:
- Amazon Polly – Natural text-to-speech synthesis
- Amazon Transcribe – Accurate speech recognition
- Amazon Translate – Multi-language support
- Text conversations
- Image generation
- Audio processing
- Embeddings for semantic search
- Complete AI workflows
| Use Case | Description |
|---|---|
| Chat Interfaces | Connect OpenWebUI or LibreChat to Bedrock models for private ChatGPT alternatives |
| Workflow Automation | Connect AI to 400+ services using N8N's visual automation platform |
| Developer Tools | AI assistance in VS Code, JetBrains IDEs, Cursor with LangChain and LlamaIndex |
| Knowledge Management | Add AI-powered writing and semantic search to Obsidian and Notion |
| Communication Bots | Deploy bots to Slack, Discord, Teams, and Telegram |
stdapi.ai is available on AWS Marketplace with commercial licensing and production-ready deployment options.
- ✅ Commercial License – Use in proprietary applications without AGPL obligations
- ✅ Hardened Container Images – Security-optimized for production workloads
- ✅ Regular Security Updates – Timely patches and vulnerability fixes
- ✅ Streamlined Deployment – Quick deployment with Terraform/OpenTofu modules
- ✅ AWS Best Practices – Following AWS Well-Architected Framework guidelines
- ✅ Enterprise Support – Professional support for deployment and configuration
Deploy stdapi.ai to your AWS account in minutes using our Terraform module:
module "stdapi_ai" {
source = "stdapi-ai/stdapi-ai/aws"
version = "~> 1.0"
}Then make your first API call:
from openai import OpenAI
client = OpenAI(
api_key="YOUR_API_KEY",
base_url="https://YOUR_DEPLOYMENT_URL/v1"
)
response = client.chat.completions.create(
model="anthropic.claude-sonnet-4-5-20250929-v1:0",
messages=[{"role": "user", "content": "Hello from AWS!"}]
)
print(response.choices[0].message.content)- Python 3.13 or higher
- uv package manager
- AWS credentials configured
-
Clone the repository
git clone https://github.com/stdapi-ai/stdapi.ai.git cd stdapi.ai -
Install dependencies
uv sync
-
Login to AWS
# Login using AWS SSO aws sso login --profile your-profile-name # Or configure your default profile aws configure sso
-
Configure the application
# Core AWS Configuration (auto-detects current region if not set) export AWS_BEDROCK_REGIONS=us-east-1 # Optional: defaults to current AWS region # S3 Storage (required for certain features like image generation, audio) export AWS_S3_BUCKET=my-dev-bucket # Create bucket in same region as AWS_BEDROCK_REGIONS # API Authentication (optional for local dev) export API_KEY=sk-dev-your-secret-key-here # Enable API documentation (helpful for development) export ENABLE_DOCS=true export ENABLE_REDOC=true # Logging Configuration export LOG_LEVEL=info # Options: info, warning, error, critical, disabled export LOG_REQUEST_PARAMS=true # Enable detailed request/response logging for debugging
-
Run locally
uv run stdapi
-
Test the API
curl http://localhost:8000/v1/models
- Follow existing code style and conventions
- Add tests for new features
- Update documentation for user-facing changes
- Ensure all tests pass before submitting PR
- Official Documentation – Complete guides and API reference
- Getting Started – Deployment and configuration
- API Reference – Detailed API documentation
- Licensing – AGPL vs Commercial licensing
This project is dual-licensed:
- AGPL-3.0-or-later – Free for open-source projects that share alike
- Commercial License – Available via AWS Marketplace for proprietary applications
The AWS Marketplace version provides full commercial rights, no source disclosure requirements, and production-ready infrastructure.
We welcome contributions! Whether it's:
- 🐛 Bug reports and fixes
- ✨ New features and enhancements
- 📖 Documentation improvements
- 💡 Ideas and suggestions
Please feel free to open issues or submit pull requests.
- 🐛 Issues: GitHub Issue Tracker
- 📖 Documentation: stdapi.ai
- 💖 Sponsor: GitHub Sponsors – Support the project's development
Sponsorship benefits include priority support, feature prioritization, dedicated development time, SLA for critical issues, and influence on the project roadmap. View sponsorship tiers →
| For Developers | For Enterprises |
|---|---|
| • Zero code changes required • Works with existing OpenAI SDKs • Simple endpoint swap • Comprehensive API compatibility |
• Data sovereignty & compliance • Cost transparency & savings • No vendor lock-in • Enterprise-grade reliability |
Get Started with AWS AI
Documentation • AWS Marketplace • GitHub Issues
Made with ❤️ for the AWS and AI community