ERNIE 5.0 Thinking Preview: benchmark results
Baidu's extended-thinking preview variant of the multimodal ERNIE 5.0 flagship. Provider: Baidu. Released 2026-01-01. Access: API.
Unified ELO 1585 ± 1, rank #305 of 1392 rated models, from 19 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA TAU-2 Bench | 83.92 | Accuracy (%) | 74.8 |
| AA CritPt | 1.43 | Accuracy (%) | 68.1 |
| AA Terminal-Bench Hard | 25 | Accuracy (%) | 66 |
| AA Humanity's Last Exam | 13.25 | Accuracy (%) | 63.1 |
| AA GPQA Diamond | 77.68 | Accuracy (%) | 62.9 |
| AA Omniscience - Science, Engineering & Mathematics | 31.24 | Accuracy (%) | 61.9 |
| Artificial Analysis Intelligence Index | 15.66 | Intelligence Index | 59.7 |
| AA Omniscience - Law | 14.43 | Accuracy (%) | 58.7 |
| AA Omniscience - Health | 22.2 | Accuracy (%) | 56.9 |
| AA Omniscience - Humanities & Social Sciences | 22.1 | Accuracy (%) | 56.8 |
| AA-Omniscience Accuracy | 22.13 | Accuracy (%) | 54.8 |
| AA Omniscience - Business | 17.39 | Accuracy (%) | 52 |
Interactive version: theaggregate.ai/model?slug=ernie-5-0-thinking-preview · How It Works · Data refreshed daily, snapshot 2026-09-05.