HellaSwag
Commonsense natural-language inference benchmark about grounded situations.
17rows
scoreprimary metric
2026-05-06sampled
Metadata
Metrics
Score, Standard error (lower is better)
| Rank | Subject | Score | Model Match | Provenance | Sampled |
|---|---|---|---|---|---|
| 1 | GPT-4.1 | 95.30 |  GPT-4.1 openai-gpt-4.1 | Imported | 2026-05-06 |
| 2 | Llama 3.1 405B | 89.20 | — | Imported | 2026-05-06 |
| 3 | falcon-180B | 89 | — | Imported | 2026-05-06 |
| 4 | DeepSeek V3 | 88.90 |  DeepSeek V3 deepseek-deepseek-chat | Imported | 2026-05-06 |
| 5 | Mixtral-8x7B-v0.1 | 86.70 | — | Imported | 2026-05-06 |
| 6 | Llama-2-70b-hf | 85.30 | — | Imported | 2026-05-06 |
| 7 | Qwen 2.5 72B | 84.80 |  Qwen2.5 72B Instruct qwen-qwen-2.5-72b-instruct | Imported | 2026-05-06 |
| 8 | Qwen2.5-Max | 83 | — | Imported | 2026-05-06 |
| 9 | Phi-3-medium-128k-instruct | 82.40 | — | Imported | 2026-05-06 |
| 10 | gemma-7b | 82.20 | — | Imported | 2026-05-06 |
| 11 | Mistral-7B-v0.1 | 81 | — | Imported | 2026-05-06 |
| 12 | Llama-2-7b | 80.70 | — | Imported | 2026-05-06 |
| 13 | Phi-3-small-8k-instruct | 77 | — | Imported | 2026-05-06 |
| 14 | Phi-3-mini-4k-instruct | 76.70 | — | Imported | 2026-05-06 |
| 15 | Yi-6B | 76.40 | — | Imported | 2026-05-06 |
| 16 | GPT-OSS 120B | 70.50 | gpt-oss-120b openai-gpt-oss-120b | Imported | 2026-05-06 |
| 17 | Phi-4 | 53.60 |  Phi 4 microsoft-phi-4 | Imported | 2026-05-06 |
No matching rows.