AI Inference Gateway and Model Router by Code Creator

Deploy a private, self-hosted AI gateway on AWS with browser-based AI chat and an OpenAI-compatible API - without building the infrastructure from scratch. AI Inference Gateway and Model Router by Code Creator delivers a fully integrated stack that is ready to serve your team from first boot.

This product is built for developers, AI teams, small businesses, and internal enterprise users who need a simple, private AI launch point on AWS. Whether you need a browser-based chat interface for internal users or an OpenAI-compatible API endpoint for applications and workflows, this AMI provides a turnkey starting point.

Example scenario: A development team that wants to test prompt chains and AI workflows against a local model before deploying to production can use the OpenAI-compatible API gateway to iterate privately, then route requests to external LLM providers when ready - all through the same unified API interface.

• Open WebUI - Browser-based chat interface supporting Ollama and OpenAI-compatible APIs
• Ollama - Local model serving with a CPU-friendly default model pre-configured and ready to use
• LiteLLM - Unified model routing proxy providing an OpenAI-compatible API layer across many LLM providers
• PostgreSQL - Integrated database for persistent data storage
• Redis - In-memory data store for caching and session management
• Nginx - Reverse proxy handling web traffic to the UI and API endpoints
• First Boot Automation - Automated provisioning that configures the full stack and sets up your instance URL on launch
• Helper Commands - Operational tooling for managing services, checking status, and common administrative tasks

Open WebUI provides the browser interface where users interact with AI models through chat. Ollama serves open models locally on the instance. LiteLLM acts as the model routing and API gateway layer, providing a unified OpenAI-compatible interface that can proxy requests to many LLM providers. Nginx sits in front of the stack as a reverse proxy, routing traffic to the appropriate services.

At first boot, the automated provisioning process configures all components, sets up the instance URL, and prepares the CPU-friendly default local model so you can start chatting and making API calls immediately.

• Private deployment - Your AI inference traffic stays on your own AWS infrastructure
• Browser-based AI chat - Give your team an intuitive chat interface without external dependencies
• OpenAI-compatible API - Connect downstream applications, scripts, and workflows through a standard API
• Multi-provider model routing - Use LiteLLM to route requests across different LLM providers through a single endpoint
• Local model serving - Run open models directly on the instance using Ollama
• Integrated data layer - PostgreSQL and Redis are pre-configured and running as part of the stack
• Operational tooling - Built-in helper commands simplify day-to-day management

Launch the AMI on your preferred EC2 instance. The first boot automation handles provisioning and configuration of all stack components. Once complete, access Open WebUI through your browser to start chatting with the default local model, or connect to the LiteLLM API endpoint to integrate with your applications. Add external LLM provider API keys through LiteLLM to route requests to additional models as needed.

Code Creator charges apply for the integrated AI inference and model routing stack, automated first boot provisioning, local model serving configuration, OpenAI-compatible API gateway setup, PostgreSQL and Redis integration, operational helper tooling, and AWS Marketplace ready AMI engineering. Standard AWS infrastructure charges for your EC2 instance apply separately.

Note: This product packages and integrates open-source components into a cohesive, pre-configured deployment on AWS. The value is in the integration, automation, and operational readiness - not the individual software components themselves.