Decision support
Claude Sonnet 3.7 vs ERNIE 4.5
Compare canonical objects directly. The useful dimensions change with the object type, so model and hardware comparisons stay separate.
Baidu
Benchmark results
Arena Text
Arena rating · Higher is better
- Claude Sonnet 3.71388Arena #184CI 1384 to 1393
- ERNIE 4.5Not reported
Runs and sources
Claude Sonnet 3.7- default138832K, thinking · 95% CI 1384 to 1393 · 38,126 votes
- default137295% CI 1368 to 1376 · 42,553 votes
ERNIE 4.5Not reported
Arena Vision
Arena rating · Higher is better
- Claude Sonnet 3.71193Arena #98CI 1177 to 1208
- ERNIE 4.5Not reported
Runs and sources
Claude Sonnet 3.7- default119332K, thinking · 95% CI 1177 to 1208 · 1,432 votes
- default117595% CI 1166 to 1185 · 4,516 votes
ERNIE 4.5Not reported
ARC-AGI-2
Best published score · Higher is better
- Claude Sonnet 3.70.9%#76 of 85
- ERNIE 4.5Not reported
Runs and sources
Claude Sonnet 3.7- default0.9%8K
- default0.7%16K
- default0.4%1K
- default0.0%
ERNIE 4.5Not reported
Source:External evaluation via Epoch AIOriginal-source terms applyFrontierMath-2025-02-28-Private
Best published score · Higher is better
- Claude Sonnet 3.74.1%#44 of 58
- ERNIE 4.5Not reported
Runs and sources
Claude Sonnet 3.7- default4.1%16K
- default3.4%32K
- default3.1%
- default3.1%64K
ERNIE 4.5Not reported
Source:Epoch AI evaluationGPQA diamond
Best published score · Higher is better
- Claude Sonnet 3.779.7%#88 of 161
- ERNIE 4.5Not reported
Runs and sources
Claude Sonnet 3.7- default79.7%64K
- default76.8%16K
- default76.8%32K
- default66.0%
ERNIE 4.5Not reported
Source:Epoch AI evaluationHumanity's Last Exam
Best published score · Higher is better
- Claude Sonnet 3.78.0%#25 of 34
- ERNIE 4.5Not reported
Runs and sources
Claude Sonnet 3.7- unknown8.0%
ERNIE 4.5Not reported
Source:External evaluation via Epoch AIOriginal-source terms applyOSWorld
Best published score · Higher is better
- Claude Sonnet 3.735.8%#8 of 10
- ERNIE 4.5Not reported
Runs and sources
Claude Sonnet 3.7- default35.8%claude-3-7-sonnet-20250219 (50 steps)
- default35.6%claude-3-7-sonnet-20250219 (100 steps)
- default27.1%claude-3-7-sonnet-20250219 (15 steps)
ERNIE 4.5Not reported
Source:OS World WebsiteOriginal-source terms applyOTIS Mock AIME 2024-2025
Best published score · Higher is better
- Claude Sonnet 3.757.8%#115 of 158
- ERNIE 4.5Not reported
Runs and sources
Claude Sonnet 3.7- default57.8%64K
- default53.3%32K
- default46.7%16K
- default21.9%
ERNIE 4.5Not reported
Source:Epoch AI evaluationSWE-bench Verified
Best published score · Higher is better
- Claude Sonnet 3.761.0%#31 of 35
- ERNIE 4.5Not reported
Runs and sources
Claude Sonnet 3.7- default61.0%
ERNIE 4.5Not reported
Source:Epoch AI evaluation
Each model's best published result is shown; "Not reported" means no matched evaluation, not a score of zero. Arena and DeepSWE ranks are the leaderboards' own; other ranks are among catalogue models with a score. Results from different harnesses or settings are not directly comparable.
Original-source terms apply to evaluations credited to their original publishers.
Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.
Arena ratings: Arena (lmarena-ai/leaderboard-dataset), licensed CC BY 4.0. Ratings come from pairwise human votes and are only comparable within one leaderboard; the interval is Arena's 95% confidence interval. Model names are matched to the Vertical Tabs catalog.
Model specifications
API pricing · USD per 1M tokens
Capabilities and modalities
Architecture and availability
A dash means the field is not reported. Provider-specific performance and pricing may differ.