SCQA (No Search) - Single-Hop Low-Visibility: leaderboard

Metric: Answer accuracy (0-1; 20 single-hop questions on facts found only in niche sources; direct prompting without web search, temperature 0, ChatGPT 5.5 judge). Source: arxiv.org. Saturation forecast: Around March 2027. 4 models tracked.

Top models

#ModelScore
1Claude Opus 4.60.55
2GLM-50.2
3DeepSeek V3.20.15
4Qwen 3 235B A22B0.05

Interactive version: theaggregate.ai/benchmark?slug=scqa-no-search-single-hop-low-visibility · How It Works · Data refreshed daily, snapshot 2026-09-25.