T5-Base: benchmark results
Provider: Other. Access: Open.
Unified ELO 1190 ± 41, rank #2640 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BoolQ | 81.4 | Accuracy (%) | 53.2 |
| Epoch AI - Superglue | 76.2 | Score | 42.9 |
| RSPC: A Benchmark for Modeling Stress and Psychiatric Conditions in Digitally Mediated Relationships using Psychiatrist Annotations | 46.1 | Macro-F1 (%) | 27.3 |
| InstructEval - Alignment (HHH) | 46.8 | Average (%, harmless/helpful/honest) | 8 |
| Open Portuguese LLM - OAB Exams | 16.13 | Accuracy (%) | 4.6 |
| Open Portuguese LLM - BLUEX | 10.15 | Accuracy (%) | 3.7 |
| Open Portuguese LLM - ENEM | 12.32 | Accuracy (%) | 3.7 |
| Open Portuguese LLM - FaQuAD NLI | 11.79 | Macro F1 (%) | 2.9 |
| Open Portuguese LLM - ASSIN2 RTE | 22.12 | Macro F1 (%) | 2.8 |
| BIRD-SQL | 12.89 | Execution Accuracy (%) | 0 |
Interactive version: theaggregate.ai/model?slug=t5-base · How It Works · Data refreshed daily, snapshot 2026-09-19.