# Natural Language Understanding on NLU tasks 21 benchmarks

Source: [Natural Language Understanding on NLU tasks 21 benchmarks](<https://www.sota2.com/research/sota/natural-language-understanding-on-nlu-tasks-21-benchmarks>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-27T07:30:42.513Z

Evaluation of general Natural Language Understanding capabilities aggregated across 21 distinct NLU tasks, reporting the average performance score.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Natural Language Understanding |
| Dataset (including any stated split) | NLU tasks 21 benchmarks |
| Metric | Avg NLU Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Avg NLU Score](<https://www.sota2.com/research/sota/natural-language-understanding-on-nlu-tasks-21-benchmarks.md?metric=Avg%20NLU%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Avg NLU Score | Result date | Report paper |
| --- | --- | --- | --- |
| PaLM | 0.747 | Not recorded | 68fa6d931e209131cea3abf2 |
| GLaM | 0.687 | Not recorded | 68fa6d931e209131cea3abf2 |
| PaLM | 0.673 | Not recorded | 68fa6d931e209131cea3abf2 |
| GPT-3 | 0.654 | Not recorded | 68fa6d931e209131cea3abf2 |
| PaLM | 0.592 | Not recorded | 68fa6d931e209131cea3abf2 |

### PaLM

Avg NLU Score: 0.747. Result ID: 69a1481c35dfcabb00e5fc94.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d931e209131cea3abf2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d931e209131cea3abf2.md>)

Recorded evaluation configuration: {"Number of Parameters":"540B","evaluation\_mode":"1-shot"}. This record does not assert independent reproduction.

### GLaM

Avg NLU Score: 0.687. Result ID: 69a1481c35dfcabb00e5fc91.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d931e209131cea3abf2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d1d1e209131cea37ddb.md>)

Recorded evaluation configuration: {"Number of Parameters":"64B/64E","evaluation\_mode":"1-shot"}. This record does not assert independent reproduction.

### PaLM

Avg NLU Score: 0.673. Result ID: 69a1481c35dfcabb00e5fc93.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d931e209131cea3abf2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d931e209131cea3abf2.md>)

Recorded evaluation configuration: {"Number of Parameters":"62B","evaluation\_mode":"1-shot"}. This record does not assert independent reproduction.

### GPT-3

Avg NLU Score: 0.654. Result ID: 69a1481c35dfcabb00e5fc90.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d931e209131cea3abf2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d1d1e209131cea37ddb.md>)

Recorded evaluation configuration: {"Number of Parameters":"175B","evaluation\_mode":"1-shot"}. This record does not assert independent reproduction.

### PaLM

Avg NLU Score: 0.592. Result ID: 69a1481c35dfcabb00e5fc92.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d931e209131cea3abf2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d931e209131cea3abf2.md>)

Recorded evaluation configuration: {"Number of Parameters":"8B","evaluation\_mode":"1-shot"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.