A suite of tools for delivering high quality LLM apps
Latitude is the end-to-end platform to build, evaluate, and refine prompts
Prompt manager
Collaborate with your team to write and iterate your prompts
Playground
Test prompts at scale before pushing them to production
Observability
Log and debug all your requests with a few lines of code
Evaluations
Run LLM-as-judge, code, or human evaluations on the generated logs
Datasets
Create or upload datasets easily to test prompts or evaluations in batch
Refiner
Automatically improve your prompts based on their evaluation results
