OmniDocBench
A document-understanding benchmark for parsing and reasoning over complex document layouts.
1 models
Higher is better.
- 1Kimi K3Moonshot AI91.1
About this benchmark
- Metric
- Document-understanding score
- Tasks
- Complex document-understanding tasks
- Authors
- Moonshot AI / OmniDocBench authors (2026)
Scores are as reported by model developers or evaluators and may use different settings (for example reasoning effort or tools). Compare with care.