LogicScore - HotpotQA Completeness: leaderboard

Metric: Completeness (%; answers whose extracted propositions link the gold answer back to a question entity; 1,939 HotpotQA distractor-setting items). Source: arxiv.org. Saturation forecast: Around May 2027. 21 models tracked.

Top models

#ModelScore
1Gemini 3 Pro78.65
2Qwen 3 30B A3B77.09
3DeepSeek V3.276.34
4Qwen 3 4B75.58
5GPT-4o75.57
6Qwen 3 8B75.36
7DeepSeek R173.91
8Qwen 3 235B A22B73.8
9O373.24
10Qwen 3 14B73.18
11GPT-5.172.61
12Qwen 3 1.7B66.17
13Qwen 3 0.6B52.16

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