adaline/gateway
The only fully local production-grade Super SDK that provides a simple, unified, and powerful interface for calling more than 200+ LLMs.
What it solves
Adaline Gateway is a TypeScript-based SDK that provides a unified interface for interacting with over 300 Large Language Models (LLMs) from various providers. It eliminates the need to write provider-specific code for every model, allowing developers to switch between models or providers without changing their core application logic.
How it works
The SDK acts as a local integration layer rather than a proxy. It uses a set of core packages and optional provider-specific packages (e.g., @adaline/openai, @adaline/anthropic) to translate generic request types into the specific schemas required by different LLM providers. It supports both complete chat responses and streaming responses, as well as word embeddings.
Who it’s for
It is designed for enterprise developers building production-grade AI applications who need a standardized way to manage multiple LLM providers while maintaining full control over their infrastructure and data privacy.
Highlights
- Unified Interface: A single API for calling 300+ LLMs across providers like OpenAI, Anthropic, Google, and AWS Bedrock.
- Production-Ready Features: Built-in support for automatic retries with exponential backoff, batching with custom queue support, and caching.
- Local Deployment: Operates entirely locally and is not a proxy, ensuring data privacy.
- Extensibility: Supports custom plugins for logging, caching, and HTTP clients, as well as plug-and-play custom providers for local models.
- Comprehensive Tooling: Includes support for tool calling, OpenTelemetry integration for instrumentation, and strong TypeScript typing.