ERNIE 4.5: benchmark results

Provider: Baidu. Released 2025-03-16. Access: API.

Unified ELO 1400 ± 1, rank #1134 of 1392 rated models, from 12 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AI for Education Visual Maths - Statistics and Probability28.57Accuracy (%)60.8
ZeroEval GPQA Diamond74GPQA Diamond Score49.8
AI for Education Visual Maths - Geometry33.85Accuracy (%)20.9
AI for Education Visual Maths - Number and Operations35.14Accuracy (%)18.2
AI for Education Visual Maths31.65Accuracy (%)14.9
AI for Education Visual Maths - Algebra30.77Accuracy (%)9.5
AI for Education Visual Maths - Measurement21.62Accuracy (%)6.8
LLM Stats (C-Eval)40.7Score (%)0
LLM Stats (CMMLU)39.8Score (%)0
LLM Stats (DROP)28.6Score (%)0
LLM Stats (MMLU-Redux)43.2Score (%)0
LLM Stats Score-13.27LLM Stats Score (conservative rating)0

Interactive version: theaggregate.ai/model?slug=ernie-4-5 · How It Works · Data refreshed daily, snapshot 2026-09-05.