SuperGLUE leaderboard
As of 6 Nov 2026, the top score on SuperGLUE is 88.9% by T5-11B (Google), ahead of T5-3B at 86.4%, across 8 model configurations with a published result (Epoch AI Benchmarking Hub). A suite of language-understanding tasks. SuperGLUE is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 6 Nov 2026)
- T5-11B holds the top SuperGLUE score at 88.9% (Epoch AI Benchmarking Hub, 6 Nov 2026).
- T5-3B is second on SuperGLUE at 86.4% (Epoch AI Benchmarking Hub, 6 Nov 2026).
- SuperGLUE: A suite of language-understanding tasks. A 2019 benchmark, saturated for current models; kept for history (Epoch AI Benchmarking Hub).
- 8 model configurations have a published SuperGLUE result on BenchLeader (BenchLeader, 6 Nov 2026).
- SuperGLUE does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
A suite of language-understanding tasks. A 2019 benchmark, saturated for current models; kept for history.
- Published by
- Epoch AI Benchmarking Hub
- Category
- Reasoning
- Index weight
- Reference only
- Models
- 8
- Data as of
- 6 Nov 2026
CC BY 4.0 — Epoch AI, ‘AI Benchmarking Hub’, epoch.ai/benchmarks. Mirrored boards credit their original publishers.
What the test looks like
A suite of language-understanding tasks.
How it is scored
Accuracy, as compiled by Epoch AI from published results.
What to keep in mind
Saturated: frontier models score near the ceiling, so it separates only older or smaller models, and results are largely self-reported by developers.
- 1T5-11B88.9%
- 2T5-3B86.4%
- 3Switch-Large84.7%
- 4T5-Large82.7%
- 5T5-Base76.2%
- 6Switch-Base73.3%
- 7InstructGPT 175B71.8%
- 8T5-Small63.3%
| # | |||
|---|---|---|---|
| 1 |  T5-11B | 88.9% | – |
| 2 | T5-3B | 86.4% | – |
| 3 | Switch-Large | 84.7% | – |
| 4 | T5-Large | 82.7% | – |
| 5 | T5-Base | 76.2% | – |
| 6 | Switch-Base | 73.3% | – |
| 7 |  InstructGPT 175B | 71.8% | – |
| 8 | T5-Small | 63.3% | – |
More reasoning benchmarks · All LLM benchmarks
Related benchmarks
Cite as: BenchLeader, “SuperGLUE leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/superglue.
SuperGLUE: questions
- What does SuperGLUE measure?
- A suite of language-understanding tasks. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on SuperGLUE?
- As of 6 Nov 2026, the highest score on SuperGLUE is 88.9% by T5-11B, ahead of T5-3B at 86.4%. Scores come from Epoch AI Benchmarking Hub and are re-read daily.
- Which AI model leads SuperGLUE?
- T5-11B leads SuperGLUE with 88.9% as of 6 Nov 2026, ahead of T5-3B at 86.4%.
- How many models have SuperGLUE results?
- 8 model configurations have a SuperGLUE result on BenchLeader, all taken from Epoch AI Benchmarking Hub.
- Who runs SuperGLUE and how often is it updated?
- SuperGLUE is published by Epoch AI Benchmarking Hub. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does SuperGLUE count toward the BenchLeader Index?
- No. SuperGLUE is shown for reference but left out of the composite index.