Back to toolsStanford NLP

Tool profile

DSPy

Stanford NLP

A framework for programming LLM pipelines with declarative modules, optimization, and evaluation-oriented workflows.

What it's used for

DSPy is used to build and tune prompt programs, retrieval pipelines, and structured multi-step AI systems with an emphasis on measurable performance.

Categories

How you access it

See whether you access this through a vendor-hosted app, managed cloud, or official client.

APIMedium complexityUsage based

Connected inference providers

Any provider

DSPy often uses hosted model providers to execute optimized prompt programs.

How you deploy or integrate it

See whether you can self-host it, deploy it in your stack, or integrate it through APIs and runtimes.

LocalMedium complexitySelf-hosted infra cost

Developer-managed runtime

Self-hosted

The framework runs inside local notebooks, services, and private infrastructure.