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

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

Top models

#ModelScore
1GPT-5.5 (xHigh)92
2GPT-5.5 (Low)88
3Claude Opus 4.8 (Adaptive Reasoning, Max Effort)88
4GPT-5.4 (xHigh)84
5Claude Opus 4.6 (Adaptive Reasoning, Max Effort)84
6Claude Opus 4.5 (Non-reasoning)84
7GPT-5.6 Sol (Max)84
8GPT-5.6 Sol (High)84
9Claude Opus 4.7 (Adaptive Reasoning, Max Effort)84
10GPT-5.6 Sol (Non-reasoning)84
11Gemini 3.1 Pro (Preview)80
12Kimi K380
13GPT-5.2 (xHigh)80
14GPT-5.5 (High)80
15GPT-5.6 Sol (xHigh)80

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