LangChain is an Observability and AI Monitoring company founded in 2022 and based in San Francisco, United States. It has raised $160M in total funding, most recently a Series B in 2025 at a $1.3B valuation.
| Date | Stage | Amount | Valuation | Lead investors |
|---|---|---|---|---|
| Oct 20, 2025 | Series B | $125M | $1.3B | IVP |
| Feb 15, 2024 | Series A | $25M | $200M |
| Sequoia Capital |
| Apr 4, 2023 | Seed | $10M | — | Benchmark |
LangChain is the leading open-source framework for building applications powered by large language models (LLMs). It provides a modular, composable architecture that lets developers chain together prompts, models, tools, and data sources into sophisticated pipelines called chains. With built-in support for agents, memory, retrieval-augmented generation (RAG), and integrations with hundreds of LLM providers and vector stores, LangChain enables developers to go from prototype to production quickly. It supports both Python and TypeScript and is the most widely adopted framework in the LLM ecosystem, used by over 50,000 companies.
LangSmith is LangChain's commercial platform for tracing, evaluating, monitoring, and deploying LLM-powered applications in production. It provides deep observability into every step of an agent's execution, allowing developers to debug, inspect, and understand LLM call behavior. LangSmith includes built-in evaluation tools to score and improve agent performance, supports A/B testing of prompts and models, and offers deployment capabilities to ship agents reliably at scale. It's designed for teams that need to move from experimental prototypes to production-grade AI systems.
LangGraph is an agent orchestration framework that enables developers to build reliable, multi-step AI agents using a graph-based architecture. Unlike linear chains, LangGraph models agent workflows as cyclic graphs where nodes represent LLM calls, tool executions, or decision points, and edges define the flow of control. This makes it ideal for building sophisticated autonomous agents that can loop, branch, and recover from errors. LangGraph handles state management, persistence, streaming, and human-in-the-loop workflows, giving developers fine-grained control over complex agent behaviors.