Spider 2.0-DBT: leaderboard
68 problems requiring navigation of dbt data pipeline projects, understanding existing SQL transformations, and generating correct queries across multi-file environments using DuckDB.
Metric: Accuracy (%). Source: spider2-sql.github.io. Saturation forecast: Around January 2027. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 41.18 | #98 |
| 2 | GPT-5.4 | 39.71 | #79 |
| 3 | Claude Sonnet 4.5 | 35.29 | #142 |
| 4 | Claude 3.7 Sonnet (20250219) | 14.7 | #201 |
| 5 | O1 Preview | 13.24 | #193 |
| 6 | GPT-4.1 | 10.29 | #248 |
| 7 | GPT-4o | 7.35 | #343 |
| 8 | O3 Mini | 4.41 | #272 |
| 9 | O3 | 2.94 | #124 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=spider-2-0-dbt · How It Works · Data refreshed daily, snapshot 2026-10-20.