Stanford NLP
Program and optimize LLM pipelines instead of hand-tuning prompts.
GitHub Stars
35.5K
Contributors
434
PyPI / Month
6.1M
What the engine gives you out of the box, in plain language.
Define tasks as modules with clear inputs and outputs, not raw prompt strings.
Optimizers search for the prompts and few-shot examples that score best.
Run the same pipeline across different models and providers.
The jobs this engine is best suited for.
Optimize the prompts behind an agent against a metric instead of guessing.
Treat prompt engineering as code you can test and version.
Move a pipeline to a new model and re-optimize without a rewrite.

Side-by-Side
Add a second or third engine and see stars, downloads, and capabilities lined up next to each other.