Benchmark
Social IQa
The first large-scale benchmark for commonsense reasoning about social situations. Contains 38,000 multiple choice questions probing emotional and social…
- Modality
- text
- Categories
- psychology, reasoning, creativity
- Openness
- unknown
- Source
- llm_stats
- Reported scores
- 9
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Gemma 2 27B | 0.537 | 2024-06-27 | evidence | |
| Gemma 2 9B | 0.534 | 2024-06-27 | evidence | |
| Gemma 3n E2B | 0.488 | 2025-06-26 | evidence | |
| Gemma 3n E2B Instructed LiteRT (Preview) | 0.488 | 2025-05-20 | evidence | |
| Gemma 3n E4B | 0.5 | 2025-06-26 | evidence | |
| Gemma 3n E4B Instructed LiteRT Preview | 0.5 | 2025-05-20 | evidence | |
| Phi 4 Mini | Microsoft | 0.725 | 2025-02-01 | evidence |
| Phi-3.5-mini-instruct | Microsoft | 0.747 | 2024-08-23 | evidence |
| Phi-3.5-MoE-instruct | Microsoft | 0.78 | 2024-08-23 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.