Agentic benchmark
DRACO leaderboard
Data Research and Analysis with Complex Operations. Every model the catalog carries a published DRACO value for, ranked by that value.
CategoryAgentic
MeasureNormalized rubric score
TasksAgentic data research and analysis tasks
DifficultyProfessional data analysis
Published byClaude Opus 5 System Card
Agentic data-analysis tasks scored against per-task rubrics at a 980K-token budget.
DRACO ranking
3 models with a published DRACO 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 | Normalized rubric score |
|---|---|---|---|
| 1 |  Claude Opus 5 | Anthropic | 88.6 |
| 2 |  Hy4 preview | Tencent | 77.2 |
| 3 |  Ling 3.0 Flash | InclusionAI | 70.4 |
Evidence key: Observed
Rows are ordered by the value Claude Opus 5 System Card 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.