# NLU Question Answering on PIQA (Accuracy)

Source: [NLU Question Answering on PIQA (Accuracy)](<https://www.sota2.com/research/sota/nlu-question-answering-on-piqa-accuracy>)

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

Data updated: 2026-06-05T07:56:22.916Z

Evaluation of model performance on the PIQA dataset for NLU Question Answering, reporting the accuracy 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 | NLU Question Answering |
| Dataset (including any stated split) | PIQA |
| Metric | PIQA Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [PIQA Accuracy](<https://www.sota2.com/research/sota/nlu-question-answering-on-piqa-accuracy.md?metric=PIQA%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | PIQA Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| FP16 | 80.25 | Not recorded | 6a22585c86173ad9c1468922 |
| FP16 | 78.94 | Not recorded | 6a22585c86173ad9c1468922 |
| SAGE-PTQ | 78.78 | Not recorded | 6a22585c86173ad9c1468922 |
| SAGE-PTQ | 77.31 | Not recorded | 6a22585c86173ad9c1468922 |
| FP16 | 75.68 | Not recorded | 6a22585c86173ad9c1468922 |
| SAGE-PTQ | 73.94 | Not recorded | 6a22585c86173ad9c1468922 |
| BiLLM | 62.5 | Not recorded | 6a22585c86173ad9c1468922 |
| BiLLM | 58.6 | Not recorded | 6a22585c86173ad9c1468922 |
| BiLLM | 57.51 | Not recorded | 6a22585c86173ad9c1468922 |

### FP16

PIQA Accuracy: 80.25. Result ID: 6a22810cfbc373592c204957.

[Reporting paper](<https://www.sota2.com/research/paper/6a22585c86173ad9c1468922.md>)

Recorded evaluation configuration: {"Backbone":"LLaMA-3-8B","Evaluation Protocol":"zero-shot"}. This record does not assert independent reproduction.

### FP16

PIQA Accuracy: 78.94. Result ID: 6a22810cfbc373592c20495a.

[Reporting paper](<https://www.sota2.com/research/paper/6a22585c86173ad9c1468922.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-3-8B","Evaluation Protocol":"zero-shot"}. This record does not assert independent reproduction.

### SAGE-PTQ

PIQA Accuracy: 78.78. Result ID: 6a22810cfbc373592c204959.

[Reporting paper](<https://www.sota2.com/research/paper/6a22585c86173ad9c1468922.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a22585c86173ad9c1468922.md>)

Recorded evaluation configuration: {"Backbone":"LLaMA-3-8B","Evaluation Protocol":"zero-shot"}. This record does not assert independent reproduction.

### SAGE-PTQ

PIQA Accuracy: 77.31. Result ID: 6a22810cfbc373592c20495c.

[Reporting paper](<https://www.sota2.com/research/paper/6a22585c86173ad9c1468922.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a22585c86173ad9c1468922.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-3-8B","Evaluation Protocol":"zero-shot"}. This record does not assert independent reproduction.

### FP16

PIQA Accuracy: 75.68. Result ID: 6a22810cfbc373592c204954.

[Reporting paper](<https://www.sota2.com/research/paper/6a22585c86173ad9c1468922.md>)

Recorded evaluation configuration: {"Backbone":"OPT-6.7B","Evaluation Protocol":"zero-shot"}. This record does not assert independent reproduction.

### SAGE-PTQ

PIQA Accuracy: 73.94. Result ID: 6a22810cfbc373592c204956.

[Reporting paper](<https://www.sota2.com/research/paper/6a22585c86173ad9c1468922.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a22585c86173ad9c1468922.md>)

Recorded evaluation configuration: {"Backbone":"OPT-6.7B","Evaluation Protocol":"zero-shot"}. This record does not assert independent reproduction.

### BiLLM

PIQA Accuracy: 62.5. Result ID: 6a22810cfbc373592c20495b.

[Reporting paper](<https://www.sota2.com/research/paper/6a22585c86173ad9c1468922.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73a81e209131cea5db8c.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-3-8B","Evaluation Protocol":"zero-shot"}. This record does not assert independent reproduction.

### BiLLM

PIQA Accuracy: 58.6. Result ID: 6a22810cfbc373592c204955.

[Reporting paper](<https://www.sota2.com/research/paper/6a22585c86173ad9c1468922.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73a81e209131cea5db8c.md>)

Recorded evaluation configuration: {"Backbone":"OPT-6.7B","Evaluation Protocol":"zero-shot"}. This record does not assert independent reproduction.

### BiLLM

PIQA Accuracy: 57.51. Result ID: 6a22810cfbc373592c204958.

[Reporting paper](<https://www.sota2.com/research/paper/6a22585c86173ad9c1468922.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73a81e209131cea5db8c.md>)

Recorded evaluation configuration: {"Backbone":"LLaMA-3-8B","Evaluation Protocol":"zero-shot"}. This record does not assert independent reproduction.

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