SQuAD v1.1: leaderboard
Extractive reading comprehension on Wikipedia passages where models identify the exact text span answering each question. All questions are guaranteed answerable.
Metric: F1 (%). Source: rajpurkar.github.io. 127 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Median Human | 91.22 |
Interactive version: theaggregate.ai/benchmark?slug=squad-v1-1 · How It Works · Data refreshed daily, snapshot 2026-10-20.