PLSQLBench - MBPP+ (Test): leaderboard

Metric: Mean test pass@1 (%; the 308 MBPP+ problems translated to PL/SQL with withheld expanded unit tests; 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 2029. 8 models tracked.

Top models

#ModelScore
1Claude Opus 4.847.73
2GPT-5.6 Sol46.75
3GPT-5.442.21
4Gemma 4 31B39.94
5GPT-5.4 Mini38.96
6Grok 4.338.31
7Gemini 2.5 Flash Lite33.44
8Llama 4 Maverick26.3

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