PaLM 2-S: benchmark results
Provider: Google. Access: API.
Aggregate Rating 1313 ± 31, rank #8490 of 10916 rated models, from 7 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BoolQ | 88.1 | Accuracy (%) | 85.1 |
| HellaSwag | 82 | Accuracy (%) | 68.4 |
| WinoGrande | 77.9 | Accuracy (%) | 67.5 |
| ARC Challenge (AI2) | 59.6 | Accuracy (%) | 53.9 |
| TriviaQA | 75.2 | Accuracy (%) | 43.4 |
| OpenBookQA | 56.2 | Accuracy (%) | 42.9 |
| Epoch AI - ECI | 94.3 | ECI Score | 11.6 |
Interactive version: theaggregate.ai/model?slug=palm-2-s · How It Works · Data refreshed daily, snapshot 2026-11-08.