Aether is an open-source AI API gateway providing a unified access layer for teams and individuals managing multiple AI services like Claude, OpenAI, and Gemini. It simplifies integration with AI models and their CLI tools by offering multi-tenancy, intelligent load balancing, cost/quota control, and health monitoring, enhancing operational oversight and resource management.

How It Works

Aether functions as a central proxy, routing requests to various AI providers through a single API endpoint. Its architecture supports intelligent load balancing across different providers and keys, multi-tenancy for user management, and granular control over costs and quotas. It also provides health monitoring for endpoints and detailed usage tracking, enabling efficient resource management and cost optimization.

Quick Start & Requirements

• Primary installation uses Docker Compose, with recommended pre-built images.
• Non-default prerequisites include PostgreSQL, Redis, and Python for key generation.
• Required environment variables cover database credentials, JWT secret, encryption key, and admin user details.
• Local development requires docker-compose.build.yml for dependencies and separate frontend setup.
• No direct links to official quick-start or docs are provided beyond the repository itself.

Highlighted Details

• Supports advanced model configurations like 1M context and 1H caching, configurable at model or key levels.
• Offers flexible load balancing modes: "Provider Priority" for specific provider preference and "Global Key Priority" for maximizing utilization across all keys.
• Free tier usage from providers is not counted towards cost quotas, as their billing multiplier is set to 0.
• Request body logging can be configured from basic info to full request/response details.

Maintenance & Community

No specific details on contributors, sponsorships, or community channels (like Discord/Slack) are present in the provided README.

Licensing & Compatibility

The project is released under the "Aether Non-Commercial Open Source License". This license strictly prohibits commercial use and integration into proprietary, closed-source systems.

Limitations & Caveats

The primary limitation is the "Non-Commercial" license, which restricts its use to non-commercial applications. The README does not detail specific technical limitations, alpha/beta status, or known bugs.