stanfordnlp/dspy: DSPy: The framework for programming—not prompting—language models

DSPy is a framework for programming language models rather than prompting them.
It lets developers build modular AI systems and optimize prompts and weights for tasks like classifiers, RAG pipelines, and agent loops.
Users write compositional Python code instead of brittle prompts, and DSPy helps teach the language model to produce higher-quality outputs.
The project directs users to its official documentation site, dspy.ai, for learning and installation details.
The repository lists related research papers on DSPy and prompt optimization, and provides a citation for the 2024 ICLR paper on compiling declarative language model calls into self-improving pipelines.

Your notes

Save this item to your library to add private notes.