# JSONSchemaBench: A Rigorous Benchmark of Structured Outputs for Language Models

*Source* · `src-geng-2025-jsonschemabench` · status: **n/a**

| field | value |
|---|---|
| kind | Preprint |
| authors | Saibo Geng, Hudson Cooper, Michał Moskal, Samuel Jenkins, Julian Berman, Nathan Ranchin, Robert West, Eric Horvitz, Harsha Nori |
| venue | arXiv |
| date | 2025-01-18 |
| url | https://arxiv.org/abs/2501.10868 |
| arxiv | 2501.10868 |
| accessed | 2026-10-24 |

## Notes

10K real-world JSON schemas; evaluates Guidance, Outlines, llama.cpp, XGrammar, OpenAI, and Gemini on efficiency, constraint coverage, and output quality, plus the official JSON Schema Test Suite.

## Backlinks (computed)

**cited by:** [Grammar-constrained decoding for valid game actions](../technique/constrained-decoding.md)

## Citation

Saibo Geng, Hudson Cooper, Michał Moskal, Samuel Jenkins, Julian Berman, Nathan Ranchin, Robert West, Eric Horvitz, Harsha Nori (2025-01-18). *JSONSchemaBench: A Rigorous Benchmark of Structured Outputs for Language Models*. arXiv https://arxiv.org/abs/2501.10868

---
Provenance: created in run `2026-10-24-techniques-llm-agents`.