SuperGLUE

Paper

Progress Over Time

Interactive timeline showing model performance evolution on SuperGLUE

State-of-the-art frontier
Open
Proprietary

SuperGLUE Leaderboard

1 models
ContextCostLicense
1
OpenAI
OpenAI
Notice missing or incorrect data?
About this benchmark

What is SuperGLUE?

SuperGLUE is a new benchmark styled after GLUE with a new set of more difficult language understanding tasks, improved resources, and a new public leaderboard. It includes 8 primary tasks: BoolQ (Boolean Questions), CB (CommitmentBank), COPA (Choice of Plausible Alternatives), MultiRC (Multi-Sentence Reading Comprehension), ReCoRD (Reading Comprehension with Commonsense Reasoning), RTE (Recognizing Textual Entailment), WiC (Word-in-Context), and WSC (Winograd Schema Challenge). The benchmark evaluates diverse language understanding capabilities including reading comprehension, commonsense reasoning, causal reasoning, coreference resolution, textual entailment, and word sense disambiguation across multiple domains.

SuperGLUE is a text benchmark evaluating models on reasoning, language, and general tasks. LLM Stats tracks 1 models on this benchmark, scored on a 0–1 scale. The current average is 0.8, with the leader at 0.8.

Compare leaders on the best AI for reasoning, best AI for language and best AI for general leaderboards.

Current leaders

o1-mini from OpenAI currently leads the SuperGLUE leaderboard with a score of 0.750 across 1 evaluated AI models.

1o1-miniOpenAI75.0%

Source paper

Title
SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systems
Authors
Alex Wang, Yada Pruksachatkun, Nikita Nangia, Amanpreet Singh, and 4 others
Published
Abstract

In the last year, new models and methods for pretraining and transfer learning have driven striking performance improvements across a range of language understanding tasks. The GLUE benchmark, introduced a little over one year ago, offers a single-number metric that summarizes progress on a diverse set of such tasks, but performance on the benchmark has recently surpassed the level of non-expert humans, suggesting limited headroom for further research. In this paper we present SuperGLUE, a new benchmark styled after GLUE with a new set of more difficult language understanding tasks, a software toolkit, and a public leaderboard. SuperGLUE is available at super.gluebenchmark.com.

FAQ

Common questions about the SuperGLUE benchmark and leaderboard.

What is the SuperGLUE benchmark?

SuperGLUE is a new benchmark styled after GLUE with a new set of more difficult language understanding tasks, improved resources, and a new public leaderboard. It includes 8 primary tasks: BoolQ (Boolean Questions), CB (CommitmentBank), COPA (Choice of Plausible Alternatives), MultiRC (Multi-Sentence Reading Comprehension), ReCoRD (Reading Comprehension with Commonsense Reasoning), RTE (Recognizing Textual Entailment), WiC (Word-in-Context), and WSC (Winograd Schema Challenge). The benchmark evaluates diverse language understanding capabilities including reading comprehension, commonsense reasoning, causal reasoning, coreference resolution, textual entailment, and word sense disambiguation across multiple domains.

What is the SuperGLUE leaderboard?

The SuperGLUE leaderboard ranks 1 AI models based on their performance on this benchmark. Currently, o1-mini by OpenAI leads with a score of 0.750. The average score across all models is 0.750.

What is the highest SuperGLUE score?

The highest SuperGLUE score is 0.750, achieved by o1-mini from OpenAI.

How many models are evaluated on SuperGLUE?

1 models have been evaluated on the SuperGLUE benchmark, with 0 verified results and 1 self-reported results.

Where can I find the SuperGLUE paper?

The SuperGLUE paper is available at https://arxiv.org/abs/1905.00537. The paper details the methodology, dataset construction, and evaluation criteria.

What categories does SuperGLUE cover?

SuperGLUE is categorized under reasoning, language, and general. The benchmark evaluates text models.

How recent are the SuperGLUE leaderboard results?

The SuperGLUE leaderboard was last updated in July 2026 and currently includes 1 evaluated models.