DS

DSPy

Framework for programming, not prompting, language models

Self-Hosted AI ★ 36.7k stars Medium setup MIT

DSPy is a framework for building language model pipelines by writing modular Python code instead of brittle prompt strings. It can compile and optimize prompts automatically against metrics for self-hosted LLM applications.

Key features

  • Declarative LLM modules
  • Automatic prompt optimization
  • Metric-driven compilation
  • Works with local models

Strengths

  • Released under the MIT license
  • Mature project with 36.7k GitHub stars
  • Written in Python

DSPy replaces

Compare DSPy

14 head-to-head comparisons.

Similar self-hosted ai apps