Ernie-4.0-8K-Preview-0518: benchmark results
Provider: Baidu. Access: API.
Aggregate Rating 1561 ± 28, rank #1531 of 10916 rated models, from 141 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenCompass Language - Sentiment Analysis - Chinese (CompassBench 2405) | 89.7 | Score (%) | 100 |
| OpenCompass Math - Arithmetic (CompassBench 2405) | 89 | Score (%) | 100 |
| OpenCompass Language - Content Summarization - English (CompassBench 2405) | 38.7 | Score (%) | 97.4 |
| OpenCompass LLM - Math (CompassBench 2405) | 63.6 | Score (%) | 94.7 |
| OpenCompass LLM - Math - Chinese (CompassBench 2405) | 63.7 | Score (%) | 94.7 |
| OpenCompass Reasoning - Abductive - Chinese (CompassBench 2405) | 90.8 | Score (%) | 94.7 |
| OpenCompass Language - Content Summarization - Chinese (CompassBench 2405) | 27.4 | Score (%) | 92.1 |
| OpenCompass Math - College - English (CompassBench 2405) | 36.8 | Score (%) | 92.1 |
| OpenCompass Reasoning - Common Sense - Chinese (CompassBench 2405) | 51.3 | Score (%) | 92.1 |
| OpenCompass Math - College - Chinese (CompassBench 2405) | 42.1 | Score (%) | 90.8 |
| OpenCompass Math - Primary School - Chinese (CompassBench 2405) | 90 | Score (%) | 90.8 |
| OpenCompass LLM - Math - English (CompassBench 2405) | 63.5 | Score (%) | 89.5 |
Interactive version: theaggregate.ai/model?slug=ernie-4-0-8k-preview-0518 · How It Works · Data refreshed daily, snapshot 2026-11-08.