Reasoning benchmark
DROP leaderboard
Discrete Reasoning Over Paragraphs. Every model the catalog carries a published DROP value for, ranked by that value.
CategoryReasoning
MeasureF1
TasksParagraph reasoning questions
DifficultyReading and numerical reasoning
Published byDeepSeek-V4 Technical Report
A reading-comprehension benchmark requiring discrete reasoning over paragraphs, reported in DeepSeek-V4 base-model evaluations.
DROP ranking
2 models with a published DROP value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | F1 |
|---|---|---|---|
| 1 | CECeleris-1 | Celeris | 81.4 |
| 2 | SPSoofi S 30B-A3B | Soofi Project | 66.5 |
Evidence key: Observed
Rows are ordered by the value DeepSeek-V4 Technical Report published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.