ERNIE-4.5-21B-A3B-PT: benchmark results
Provider: Baidu. Released 2025-06-30. Access: Open.
Aggregate Rating 1466 ± 23, rank #2983 of 10916 rated models, from 257 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Bosnian NLU - Multi Wiki QA BS | 69.92 | Character F1 (%) | 88.1 |
| EuroEval Albanian NLU - MultiWikiQA SQ | 62.72 | Character F1 (%) | 87.9 |
| Open Portuguese LLM - ENEM | 74.88 | Accuracy (%) | 87.7 |
| EuroEval Bosnian | 54.03 | Average Score (%) | 87.1 |
| Open Portuguese LLM - OAB Exams | 54.76 | Accuracy (%) | 87.1 |
| Open Portuguese LLM - BLUEX | 65.37 | Accuracy (%) | 86.5 |
| EuroEval Bulgarian NLU - MultiWikiQA BG | 73.87 | Character F1 (%) | 85.8 |
| EuroEval Catalan NLU - MultiWikiQA CA | 73.32 | Character F1 (%) | 85.3 |
| EuroEval English NLU - SQuAD | 84.71 | Character F1 (%) | 85 |
| Open Portuguese LLM - ASSIN2 RTE | 92.85 | Macro F1 (%) | 84.4 |
| EuroEval French NLU - FQuAD | 71.42 | Character F1 (%) | 83.1 |
| EuroEval Romanian NLU - MultiWikiQA RO | 71.05 | Character F1 (%) | 83 |
Interactive version: theaggregate.ai/model?slug=ernie-4-5-21b-a3b-pt · How It Works · Data refreshed daily, snapshot 2026-11-08.