PIQA
Physical Interaction Question Answering benchmark for physical commonsense reasoning.
16rows
scoreprimary metric
2026-05-06sampled
Metadata
Metrics
Score, Standard error (lower is better)
| Rank | Subject | Score | Model Match | Provenance | Sampled |
|---|---|---|---|---|---|
| 1 | gpt-4o-mini-2024-07-18 | 88.70 |  GPT-4o-mini (2024-07-18) openai-gpt-4o-mini-2024-07-18 | Imported | 2026-05-06 |
| 2 | Phi-3-mini-4k-instruct | 88.60 | — | Imported | 2026-05-06 |
| 3 | Gemini 1.5 Flash | 87.50 | — | Imported | 2026-05-06 |
| 4 | Llama 3.1 405B | 85.90 | — | Imported | 2026-05-06 |
| 5 | falcon-180B | 84.90 | — | Imported | 2026-05-06 |
| 6 | DeepSeek V3 | 84.70 |  DeepSeek V3 deepseek-deepseek-chat | Imported | 2026-05-06 |
| 7 | Gemma 3 27B | 83.70 |  Gemma 3 27B google-gemma-3-27b-it | Imported | 2026-05-06 |
| 8 | Mixtral-8x7B-v0.1 | 83.60 | — | Imported | 2026-05-06 |
| 9 | Mistral Large | 83.50 |  Mistral Large mistralai-mistral-large | Imported | 2026-05-06 |
| 10 | Mistral-7B-v0.1 | 83 | — | Imported | 2026-05-06 |
| 11 | Llama-2-70b-hf | 82.80 | — | Imported | 2026-05-06 |
| 12 | Qwen 2.5 72B | 82.60 |  Qwen2.5 72B Instruct qwen-qwen-2.5-72b-instruct | Imported | 2026-05-06 |
| 13 | Llama-2-7b | 81.90 | — | Imported | 2026-05-06 |
| 14 | gemma-7b | 81.20 | — | Imported | 2026-05-06 |
| 15 | Qwen 3 235B | 79.90 | Qwen3 235B A22B qwen-qwen3-235b-a22b | Imported | 2026-05-06 |
| 16 | GPT-OSS 120B | 76.70 | gpt-oss-120b openai-gpt-oss-120b | Imported | 2026-05-06 |
No matching rows.