# Commonsense Reasoning and Knowledge on ARC-C, BoolQ, Lambada, PIQA, Winogrande

Source: [Commonsense Reasoning and Knowledge on ARC-C, BoolQ, Lambada, PIQA, Winogrande](<https://www.sota2.com/research/sota/commonsense-reasoning-and-knowledge-on-arc-c-boolq-lambada-piqa-winogrande>)

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

Data updated: 2026-03-31T09:10:41.469Z

Evaluation of model performance across five specific commonsense reasoning benchmarks: ARC-Challenge (ARC-C), BoolQ, Lambada, PIQA, and Winogrande, reporting individual scores and the overall average.

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 | Commonsense Reasoning and Knowledge |
| Dataset (including any stated split) | lm-eval ARC-C, BoolQ, Lambada, PIQA, Winogrande |
| Metric | ARC-C Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [ARC-C Accuracy](<https://www.sota2.com/research/sota/commonsense-reasoning-and-knowledge-on-arc-c-boolq-lambada-piqa-winogrande.md?metric=ARC-C%20Accuracy>): Higher is better; unit: PERCENT.

- [BoolQ Accuracy](<https://www.sota2.com/research/sota/commonsense-reasoning-and-knowledge-on-arc-c-boolq-lambada-piqa-winogrande.md?metric=BoolQ%20Accuracy>): Higher is better; unit: PERCENT.

- [Lambada Accuracy](<https://www.sota2.com/research/sota/commonsense-reasoning-and-knowledge-on-arc-c-boolq-lambada-piqa-winogrande.md?metric=Lambada%20Accuracy>): Higher is better; unit: PERCENT.

- [PIQA Accuracy](<https://www.sota2.com/research/sota/commonsense-reasoning-and-knowledge-on-arc-c-boolq-lambada-piqa-winogrande.md?metric=PIQA%20Accuracy>): Higher is better; unit: PERCENT.

- [Winogrande Accuracy](<https://www.sota2.com/research/sota/commonsense-reasoning-and-knowledge-on-arc-c-boolq-lambada-piqa-winogrande.md?metric=Winogrande%20Accuracy>): Higher is better; unit: PERCENT.

- [Average Accuracy](<https://www.sota2.com/research/sota/commonsense-reasoning-and-knowledge-on-arc-c-boolq-lambada-piqa-winogrande.md?metric=Average%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | ARC-C Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| FP16 | 53.58 | Not recorded | 68fa7a6a1e209131cea83140 |
| MicroMix | 52.22 | Not recorded | 68fa7a6a1e209131cea83140 |
| FlatQuant | 50.17 | Not recorded | 68fa7a6a1e209131cea83140 |
| QUIK | 49.66 | Not recorded | 68fa7a6a1e209131cea83140 |
| INT6 | 49.15 | Not recorded | 68fa7a6a1e209131cea83140 |
| Atom | 47.95 | Not recorded | 68fa7a6a1e209131cea83140 |
| AMXFP4 | 46.84 | Not recorded | 68fa7a6a1e209131cea83140 |
| QuaRot | 44.88 | Not recorded | 68fa7a6a1e209131cea83140 |

### FP16

ARC-C Accuracy: 53.58. Result ID: 69cb8e9347f34ce924b41108.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a6a1e209131cea83140.md>)

Recorded evaluation configuration: {"Avg. Bits":"16.00","Base Model":"Llama3.1-8B","Evaluation Protocol":"0-shot","Quantization":"None"}. This record does not assert independent reproduction.

### MicroMix

ARC-C Accuracy: 52.22. Result ID: 69cb8e9347f34ce924b4110f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a6a1e209131cea83140.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a6a1e209131cea83140.md>)

Recorded evaluation configuration: {"Avg. Bits":"5.51","Base Model":"Llama3.1-8B","Evaluation Protocol":"0-shot","KV Cache Quantization":"Asymmetric INT4 (group size 64)"}. This record does not assert independent reproduction.

### FlatQuant

ARC-C Accuracy: 50.17. Result ID: 69cb8e9347f34ce924b4110c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a6a1e209131cea83140.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a8b1e209131cea83cda.md>)

Recorded evaluation configuration: {"Avg. Bits":"4.19","Base Model":"Llama3.1-8B","Evaluation Protocol":"0-shot","KV Cache Quantization":"Asymmetric INT4 (group size 64)"}. This record does not assert independent reproduction.

### QUIK

ARC-C Accuracy: 49.66. Result ID: 69cb8e9347f34ce924b4110a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a6a1e209131cea83140.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71141e209131cea4f205.md>)

Recorded evaluation configuration: {"Avg. Bits":"5.95","Base Model":"Llama3.1-8B","Evaluation Protocol":"0-shot","KV Cache Quantization":"Asymmetric INT4 (group size 64)"}. This record does not assert independent reproduction.

### INT6

ARC-C Accuracy: 49.15. Result ID: 69cb8e9347f34ce924b4110e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a6a1e209131cea83140.md>)

Recorded evaluation configuration: {"Avg. Bits":"6.00","Base Model":"Llama3.1-8B","Evaluation Protocol":"0-shot","KV Cache Quantization":"Asymmetric INT4 (group size 64)"}. This record does not assert independent reproduction.

### Atom

ARC-C Accuracy: 47.95. Result ID: 69cb8e9347f34ce924b4110b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a6a1e209131cea83140.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72f11e209131cea5970b.md>)

Recorded evaluation configuration: {"Avg. Bits":"4.25","Base Model":"Llama3.1-8B","Evaluation Protocol":"0-shot","KV Cache Quantization":"Asymmetric INT4 (group size 64)"}. This record does not assert independent reproduction.

### AMXFP4

ARC-C Accuracy: 46.84. Result ID: 69cb8e9347f34ce924b4110d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a6a1e209131cea83140.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78f21e209131cea7b600.md>)

[Implementation](<https://www.sota2.com/aiha-lab/MX-QLLM.git>)

Recorded evaluation configuration: {"Avg. Bits":"5.00","Base Model":"Llama3.1-8B","Evaluation Protocol":"0-shot","KV Cache Quantization":"Asymmetric INT4 (group size 64)"}. This record does not assert independent reproduction.

### QuaRot

ARC-C Accuracy: 44.88. Result ID: 69cb8e9347f34ce924b41109.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a6a1e209131cea83140.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756a1e209131cea67b8c.md>)

[Implementation](<https://github.com/spcl/QuaRot>)

Recorded evaluation configuration: {"Avg. Bits":"4.12","Base Model":"Llama3.1-8B","Evaluation Protocol":"0-shot","KV Cache Quantization":"Asymmetric INT4 (group size 64)"}. This record does not assert independent reproduction.

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