PLSQLBench - Spider2 Single-Turn (Test): leaderboard

Metric: Mean test pass@1 (%; the 103 held-out, database-disjoint single-turn tasks over enterprise-style Spider 2 schemas; direct generation of Oracle PL/SQL program units at temperature 0, executed on Oracle Autonomous Database 23ai; per-task fraction of unit tests passed by one solution, averaged over tasks). Source: arxiv.org. Saturation forecast: Around March 2027. 8 models tracked.

Top models

#ModelScore
1GPT-5.479.47
2Claude Opus 4.876.34
3Gemma 4 31B76.26
4GPT-5.6 Sol71.57
5GPT-5.4 Mini70.9
6Grok 4.366.98
7Llama 4 Maverick61.24
8Gemini 2.5 Flash Lite60.94

Interactive version: theaggregate.ai/benchmark?slug=plsqlbench-spider2-single-turn-test · How It Works · Data refreshed daily, snapshot 2026-09-26.