ERNIE-4.5-0.3B-PT: benchmark results
Provider: Baidu. Released 2025-06-30. Access: Open.
Aggregate Rating 1208 ± 28, rank #9571 of 10916 rated models, from 257 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Icelandic NLU - ScaLA IS | 2.28 | Matthews correlation coefficient (%) | 45.3 |
| EuroEval Catalan NLU - ScaLA CA | 1.47 | Matthews correlation coefficient (%) | 43.1 |
| EuroEval Lithuanian Common Sense Reasoning | 1.29 | Common Sense Reasoning Average Score (%) | 33.2 |
| EuroEval Norwegian Knowledge - Idioms NO | 0.51 | MCC (x100) | 30 |
| EuroEval Croatian Common Sense Reasoning | 2.83 | Common Sense Reasoning Average Score (%) | 29.5 |
| EuroEval Albanian Common Sense Reasoning | 1.22 | Common Sense Reasoning Average Score (%) | 29.1 |
| EuroEval Latvian Common Sense Reasoning | 5.32 | Common Sense Reasoning Average Score (%) | 28.4 |
| EuroEval Slovene Common Sense Reasoning | 1.43 | Common Sense Reasoning Average Score (%) | 26.9 |
| EuroEval Estonian Common Sense Reasoning | 1.59 | Common Sense Reasoning Average Score (%) | 26.3 |
| EuroEval Bulgarian Common Sense Reasoning | 2.28 | Common Sense Reasoning Average Score (%) | 24.7 |
| EuroEval Bulgarian NLU - MultiWikiQA BG | 27.12 | Character F1 (%) | 24.6 |
| EuroEval Greek Common Sense Reasoning | 1.89 | Common Sense Reasoning Average Score (%) | 24.2 |
Interactive version: theaggregate.ai/model?slug=ernie-4-5-0-3b-pt · How It Works · Data refreshed daily, snapshot 2026-11-08.