GLUE CoLA: leaderboard
Corpus of Linguistic Acceptability (Warstadt et al., 2019), the GLUE task of judging whether English sentences taken from linguistics publications are grammatically acceptable. Archived Papers With Code table of scores as each paper reported them, labelled accuracy rather than GLUE’s official Matthews correlation; higher is better.
Metric: Accuracy. Source: huggingface.co. Saturation forecast: Rough model projection: around 2026. 42 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | T5 (11B) | 70.8 | #1631 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=glue-cola · How It Works · Data refreshed daily, snapshot 2026-11-05.