CommandCodeAI/BaseAI

BaseAI uses a "pipe" abstraction—declarative TypeScript configurations that define LLM behavior, tools, and memory—deployed through Langbase's serverless infrastructure with support for streaming responses and parallel tool execution. It integrates with multiple LLM providers (OpenAI, Anthropic, Cohere, etc.) via a unified API and includes built-in RAG capabilities through its memory system. The project is now archived in favor of Langbase AI Primitives, reflecting a shift away from framework-based approaches toward composable APIs for faster iteration.

1,215 stars.