A minimalist LLM framework

ZeroGraph TypeScript: 100-line minimalist LLM framework. Let Agents build Agents!

LLM framework integrations for Schema Registry (LangChain, LlamaIndex, vLLM)

Typescript bindings for langchain

A config-driven LLM framework built on top of PocketFlow

Typescript-first LLM framework with static type inference, testability, and composability.

<p align="center"> <img height="100" width="100" alt="LlamaIndex logo" src="https://ts.llamaindex.ai/square.svg" /> </p> <h1 align="center">LlamaIndex.TS</h1> <h3 align="center"> Data framework for your LLM application. </h3>

Core types, interfaces and classes for OrkaJS - TypeScript LLM framework

Simple, opinionated, JSON-typed, and traced LLM framework for TypeScript.

Core package for PocketFlow.js - A minimal LLM framework

The zero-boilerplate LLM framework for humans.

OpenAI plugin for LiveKit Node Agents

Client library to connect to the LangSmith Observability and Evaluation Platform.

All-In-One LLM Framework - Multi-provider LLM integration with auto-fallback, priority management, multimodal support, and XML-based tool calling

LangGraph

Mastra is a framework for building AI-powered applications and agents with a modern TypeScript stack.

Code Mode: use LLMs to generate executable code that performs tool calls

A minimalist LLM framework for building AI agents and workflows with JSON configuration

OpenAI-compatible function definitions and executor for SwarmSync.AI — works with AutoGen, OpenAI, and any function-calling LLM framework.

Unified LLM framework with prompt templates, model providers, and tools