PiSAR - OPeRA: leaderboard

Metric: Semantic similarity (mean cosine similarity between OpenAI text-embedding-3-small embeddings of the predicted and the gold rationale; 119 held-out OPeRA online-shopping traces). Source: arxiv.org. Saturation forecast: Around April 2028. 5 models tracked.

Top models

#ModelScore
1GPT-5.50.41
2Claude Opus 4.70.34

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