ERNIE-4.0: benchmark results
Provider: Baidu. Access: API.
Aggregate Rating 1554 ± 22, rank #1641 of 10916 rated models, from 69 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenCompass Language - Content Critique - Chinese (CompassBench 2404) | 56 | Score (%) | 100 |
| OpenCompass Reasoning - Abductive - English (CompassBench 2404) | 92 | Score (%) | 100 |
| OpenCompass Code - MBPP - English (CompassBench 2404) | 72.4 | Score (%) | 93.3 |
| OpenCompass Knowledge - Common Sense - Chinese (CompassBench 2404) | 76 | Score (%) | 93.3 |
| OpenCompass Knowledge - Humanities - Chinese (CompassBench 2404) | 54.7 | Score (%) | 93.3 |
| OpenCompass Knowledge - Natural Science - Chinese (CompassBench 2404) | 56 | Score (%) | 93.3 |
| OpenCompass Math - Primary School - English (CompassBench 2404) | 83.3 | Score (%) | 92 |
| OpenCompass Agent - T-Eval - English (CompassBench 2404) | 80.1 | Score (%) | 90 |
| OpenCompass LLM - Knowledge - Chinese (CompassBench 2404) | 63.4 | Score (%) | 90 |
| OpenCompass Math - Primary School - Chinese (CompassBench 2404) | 86.3 | Score (%) | 89 |
| OpenCompass Language - Traditional Cultural Understanding - Chinese (CompassBench 2404) | 59.3 | Score (%) | 88.3 |
| OpenCompass Agent - T-Eval - Chinese (CompassBench 2404) | 75.2 | Score (%) | 86.7 |
Interactive version: theaggregate.ai/model?slug=ernie-4-0 · How It Works · Data refreshed daily, snapshot 2026-11-08.