CONTINUUM Books
Evaluating Language Model Agents
Principles, metrics and practice
Monograph · multiple authors
Evaluating Language Model Agents
Principles, metrics and practice
Published 30 September 2026 · 312 pages · English
ISBN 978-0-00-100001-2 · DOI 10.xxxx/ctm.bk.2026.0001
Sample book. Title, authors and content are fictional demonstration material.
About this book
A practical and rigorous guide to measuring what agents actually do. The book develops a measurement framework for tool-using language agents, covers reliability, cost and safety metrics, and shows how to design benchmarks, report variance and avoid the most common evaluation mistakes. Worked examples and code accompany every chapter.
Keywords agents · evaluation · metrics · benchmark design
Read or download
evaluating-language-model-agents.pdf
evaluating-language-model-agents.epub
evaluating-language-model-agents.html
Contents
- 1Why agent evaluation is differentpp. 1–24PDF
- 2A measurement frameworkpp. 25–62PDF
- 3Success, reliability and costpp. 63–104PDF
- 4Designing benchmark taskspp. 105–150PDF
- 5Statistics for stochastic systemspp. 151–198PDF
- 6Human and model-based judgingpp. 199–236PDF
- 7Safety and robustness evaluationpp. 237–278PDF
- 8Reporting and reproducibilitypp. 279–312PDF
About the authors
- Lena Hartwell · Open Systems Lab, Germany
Researches tool-using language agents and the interfaces between models and external software.
- Kofi Mensah · Lakeshore Centre for Machine Learning, Ghana
Interested in benchmark design, measurement validity and statistical practice in machine learning.
How to cite
Hartwell, L. & Mensah, K. (2026). Evaluating Language Model Agents: Principles, metrics and practice (1st edition). CONTINUUM Books. doi:10.xxxx/ctm.bk.2026.0001