- Home
- Guides and documents
- OpenAI Cookbook: Evaluation Examples
OpenAI Cookbook: Evaluation Examples
Worked examples for evaluating prompts, RAG and SQL generation
What it is
The evaluation folder of the OpenAI Cookbook. It includes a written guide to the evaluation flywheel for making prompts hold up across inputs, a getting-started notebook for OpenAI Evals, and examples for scoring summaries, SQL generation and RAG pipelines, plus notes on moving from OpenAI Evals to promptfoo.
Good for
- Set up an evaluation loop for a prompt
- Score summarisation quality
- Evaluate text-to-SQL output
What you need
OpenAI API key for the notebooks
Worth knowingTied to OpenAI tooling; several notebooks predate current models and APIs.
Sources
- https://github.com/openai/openai-cookbook/tree/main/examples/evaluation (project README, read 11 Nov 2026)
- https://api.github.com/repos/openai/openai-cookbook (GitHub API, read 13 Nov 2026)
The summary is ours, written from the README. Spotted an error? See how we work.