VECTOR WIREAI INTELLIGENCE
⌘K
PKT
Refresh Models —Deals —Regulatory —Sources —

Questions this page answers

Which is better, DeepSeek-V3.2-Exp-Base or ERNIE 4.5?
Across 15 shared benchmarks, DeepSeek-V3.2-Exp-Base scores higher on 14 and ERNIE 4.5 on 1. The widest gap is MMLU-Pro, where DeepSeek-V3.2-Exp-Base scores 63.3 against 16.

DeepSeek-V3.2-Exp-Base vs ERNIE 4.5

Across 15 shared benchmarks, DeepSeek-V3.2-Exp-Base scores higher on 14 and ERNIE 4.5 on 1. The widest gap is MMLU-Pro, where DeepSeek-V3.2-Exp-Base scores 63.3 against 16.

DeepSeekvsBaidu15 shared benchmarks14–1 head-to-head
BenchmarkDeepSeek-V3.2-Exp-BaseERNIE 4.5
arc_challenge95.540.6
bbh88.730.4
C-Eval9140.7
cmmlu88.939.8
DROP86.628.6
GPQA Diamond5274
GSM8K91.125.2
hellaswag89.433
HumanEval+67.725
MBPP+69.840.2
mmlu_redux90.443.2
MMLU-Pro63.316
PIQA85.155.2
simpleqa271.8
winogrande85.651.3

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.