# Commonsense Reasoning on Hellaswag Multilingual (test)

Source: [Commonsense Reasoning on Hellaswag Multilingual (test)](<https://www.sota2.com/research/sota/commonsense-reasoning-on-hellaswag-multilingual-test>)

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

Data updated: 2026-02-26T05:30:51.441Z

Evaluation of model performance on the multilingual test split of the Hellaswag dataset for commonsense reasoning tasks, measured by Accuracy.

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 |
| Dataset (including any stated split) | Hellaswag Multilingual (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/commonsense-reasoning-on-hellaswag-multilingual-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Mistral-3.2-24B | 83.1 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| Qwen-3-32B | 79.6 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| Qwen-3-30B-A3B | 78.5 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| Qwen-3-14B | 76.4 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| Gemma-3-27B | 75.6 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| Gemma-3-12B | 73.5 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| Llama-3.3-70B | 73.4 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| Apertus-70B | 67.4 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| EuroLLM-22B (old) | 65 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| EuroLLM-22B (new) | 62.3 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| EuroLLM-9B (old) | 53.9 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| Apertus-8B | 50.2 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| EuroLLM-9B (new) | 49.1 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| OLMo-3.1-32B | 47.4 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| Llama-3.1-8B | 37.4 | Not recorded | 698575d8c97a0b3e3b5d5e8c |
| OLMo-3-7B | 30.1 | Not recorded | 698575d8c97a0b3e3b5d5e8c |

### Mistral-3.2-24B

Accuracy: 83.1. Result ID: 699647a15698665ec48b9a63.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d61e209131cea4d909.md>)

[Implementation](<https://github.com/mistralai/mistral-src>)

Recorded evaluation configuration: {"Openness":"Open-weights","Region":"European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### Qwen-3-32B

Accuracy: 79.6. Result ID: 699647a15698665ec48b9a69.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Openness":"Open-weights","Region":"Non-European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### Qwen-3-30B-A3B

Accuracy: 78.5. Result ID: 699647a15698665ec48b9a6a.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Openness":"Open-weights","Region":"Non-European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### Qwen-3-14B

Accuracy: 76.4. Result ID: 699647a15698665ec48b9a68.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Openness":"Open-weights","Region":"Non-European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### Gemma-3-27B

Accuracy: 75.6. Result ID: 699647a15698665ec48b9a67.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Openness":"Open-weights","Region":"Non-European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### Gemma-3-12B

Accuracy: 73.5. Result ID: 699647a15698665ec48b9a66.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Openness":"Open-weights","Region":"Non-European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### Llama-3.3-70B

Accuracy: 73.4. Result ID: 699647a15698665ec48b9a65.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Openness":"Open-weights","Region":"Non-European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### Apertus-70B

Accuracy: 67.4. Result ID: 699647a15698665ec48b9a60.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b6c1e209131cea885c7.md>)

Recorded evaluation configuration: {"Openness":"Fully-open","Region":"European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### EuroLLM-22B (old)

Accuracy: 65. Result ID: 699647a15698665ec48b9a5d.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

Recorded evaluation configuration: {"Openness":"Fully-open","Region":"European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### EuroLLM-22B (new)

Accuracy: 62.3. Result ID: 699647a15698665ec48b9a5e.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

Recorded evaluation configuration: {"Openness":"Fully-open","Region":"European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### EuroLLM-9B (old)

Accuracy: 53.9. Result ID: 699647a15698665ec48b9a5b.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

Recorded evaluation configuration: {"Openness":"Fully-open","Region":"European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### Apertus-8B

Accuracy: 50.2. Result ID: 699647a15698665ec48b9a5f.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b6c1e209131cea885c7.md>)

Recorded evaluation configuration: {"Openness":"Fully-open","Region":"European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### EuroLLM-9B (new)

Accuracy: 49.1. Result ID: 699647a15698665ec48b9a5c.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

Recorded evaluation configuration: {"Openness":"Fully-open","Region":"European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### OLMo-3.1-32B

Accuracy: 47.4. Result ID: 699647a15698665ec48b9a62.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/694238e845614693628503e0.md>)

Recorded evaluation configuration: {"Openness":"Fully-open","Region":"Non-European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### Llama-3.1-8B

Accuracy: 37.4. Result ID: 699647a15698665ec48b9a64.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Openness":"Open-weights","Region":"Non-European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

### OLMo-3-7B

Accuracy: 30.1. Result ID: 699647a15698665ec48b9a61.

[Reporting paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/694238e845614693628503e0.md>)

Recorded evaluation configuration: {"Openness":"Fully-open","Region":"Non-European","Tuning":"Instruction-tuned"}. This record does not assert independent reproduction.

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