PARROT - MySQL Dialect Compatibility: leaderboard
The "MySQL Dialect Compatibility" subtask of PARROT. PARROT SQL-dialect benchmark (code4db) — tests whether models translate and execute SQL correctly across database dialects, scoring dialect compatibility (including MySQL and Oracle) and cross-dialect result consistency.
Metric: Accuracy (%). Source: code4db.github.io. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek R1 Distill Qwen 32B | 58.82 | #655 |
| 2 | GPT-4o | 50 | #343 |
| 3 | Claude 3.7 Sonnet | 44.12 | #249 |
| 4 | O3 Mini | 8.82 | #272 |
Interactive version: theaggregate.ai/benchmark?slug=parrot-mysql-dialect-compatibility · How It Works · Data refreshed daily, snapshot 2026-10-20.