math benchmark
DROP
DROP (Discrete Reasoning Over Paragraphs) is a reading comprehension benchmark requiring discrete reasoning over paragraph content. It contains crowdsourced, adversarially-created questions that require resolving references and performing discrete operations like addition, counting, or sorting, demanding comprehensive paragraph understanding beyond paraphrase-and-entity-typing shortcuts.
Updated Aug 11, 2026
Models30
Model coverage30
MetricScore
EvidenceB
DROP Ranking
Higher score ranks better on this benchmark.
DROP Score Distribution
A closer view of the leading scores on this benchmark.
DROP Highlights
The leading models and scores on this benchmark.
What is DROP?
What DROP measures and how its scores work.
DROP (Discrete Reasoning Over Paragraphs) is a reading comprehension benchmark requiring discrete reasoning over paragraph content. It contains crowdsourced, adversarially-created questions that require resolving references and performing discrete operations like addition, counting, or sorting, demanding comprehensive paragraph understanding beyond paraphrase-and-entity-typing shortcuts.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- DROP
- Modality
- text
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- drop|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about DROP.
Which model scores highest on DROP?
DeepSeek-V3 is currently ranked first with 91.6%.
What does DROP measure?
DROP (Discrete Reasoning Over Paragraphs) is a reading comprehension benchmark requiring discrete reasoning over paragraph content. It contains crowdsourced, adversarially-created questions that require resolving references and performing discrete operations like addition, counting, or sorting, demanding comprehensive paragraph understanding beyond paraphrase-and-entity-typing shortcuts.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
30 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.