# What is DSPy? Definition and meaning

A framework that expresses an LLM pipeline as modules and optimizes its prompts and weights against a metric.

## Definition

**DSPy**: A framework that expresses an LLM pipeline as modules and optimizes its prompts and weights against a metric.

In short: Programming rather than prompting language models.

## When it is used

Systematically improving prompts and pipelines instead of hand-tuning them.

## Where it fits

DSPy belongs to **Frameworks** in the [Agentic AI](/topics/agentic-ai) topic. These are the other terms in the same group:

- **[LangGraph](/glossary/langgraph)**: A library for building stateful, long-running agents as graphs of nodes with checkpoints and human-in-the-loop steps.
- **[CrewAI](/glossary/crewai)**: A framework in which a crew of agents with roles and goals collaborate on tasks.
- **[Pydantic AI](/glossary/pydantic-ai)**: An agent framework from the Pydantic team that validates inputs and outputs with typed models.
- **[Mastra](/glossary/mastra)**: A TypeScript framework for building agents, workflows and retrieval pipelines.
- **[smolagents](/glossary/smolagents)**: A small library where agents write and run code actions to solve tasks.
- **[LlamaIndex](/glossary/llamaindex)**: A framework for connecting language models to data through indexing, retrieval and agent workflows.

## Learn more

The full guide to **[DSPy](/tools/dspy)** explains how it works, when to use it and how it compares with the alternatives, with examples. The [glossary](/glossary) lists every term on the site from A to Z.

Previous term: [Dropout](/glossary/dropout) · Next term: [DuckDB](/glossary/duckdb)
