Who leads the GLUE benchmark?

DeBERTa-v3-large currently leads GLUE with a score of 91.37 on Avg Score.

What is the state-of-the-art score on GLUE?

The state-of-the-art result on GLUE is 91.37 (Avg Score), achieved by DeBERTa-v3-large as of 2023.

How many models are tracked on GLUE?

Codesota tracks 3 models on GLUE.

When was the GLUE leaderboard last updated?

The GLUE leaderboard on Codesota includes results through 2023, with the earliest tracked result from 2019.

Codesota · Benchmark · GLUEHome/Leaderboards/GLUE

Unknown

GLUE.

Name: GLUE Benchmark Results
Creator: Unknown
Published: 2019-01-01
License: https://creativecommons.org/licenses/by/4.0/

General Language Understanding Evaluation for masked language models

Paper ↗Leaderboard ↓

§ 01 · SOTA history

Year over year.

§ 02 · Leaderboard

Results by metric.

Only 3 models on this benchmark

Help build the community leaderboard — submit your model results.

Found a wrong score or missing run?

Use row edits to send a sourced correction into moderation.

Add / edit result ↗Report issue ↗

Avg Score

Avg Score is the reported evaluation metric for GLUE. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Avg Scoreverifiedpapervendorcommunityunverified

Rank	Model	Trust	Score	Year	Links	Fix
01	DeBERTa-v3-large DeBERTa-v3-large GLUE test average. From DeBERTaV3 paper Table 2 and GLUE leaderboard.	verified	91.37	2023	Source ↗	Looks wrong?
02	ALBERT-xxlarge-v2 ALBERT-xxlarge-v2 GLUE test average (single model, not ensemble). From ALBERT paper Table 1.	verified	89.4	2020	Source ↗	Looks wrong?
03	RoBERTa-large RoBERTa-large GLUE test average (single model). From original RoBERTa paper and GLUE leaderboard.	verified	88.5	2019	Source ↗	Looks wrong?

§ 04 · Submit a result

Add to the leaderboard.

← Back to Leaderboards