AA Omniscience - Software Engineering (SWE) - Java — leaderboard

Metric: Accuracy (%). Source: artificialanalysis.ai. 449 models tracked.

Top models

#ModelScore
1GPT-5.3 Codex (xHigh)73
2GPT-5.6 Sol (Max)72
3GPT-5.5 (xHigh)71
4GPT-5.4 (xHigh)71
5GPT-5.6 Sol (xHigh)69
6GPT-5.5 (Medium)69
7GPT-5.6 Sol (High)68
8GPT-5.5 (High)67
9GPT-5.5 (Low)66
10GPT-5.4 (Low)65
11GPT-5.6 Sol (Medium)65
12GPT-5.6 Sol (Low)64
13Claude Opus 4.7 (Adaptive Reasoning, Max Effort)62
14Claude Opus 4.6 (Adaptive Reasoning, Max Effort)58
15Grok 4.5 (High)58

Interactive version: theaggregate.ai/benchmark?slug=aa-omniscience-software-engineering-swe-java · How the rankings work · Data refreshed daily, snapshot 2026-07-22.