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

Questions this page answers

Which is better, DeepSeek-V3.2-Exp-Base or Mistral 7B?
Across 12 shared benchmarks, DeepSeek-V3.2-Exp-Base scores higher on 12 and Mistral 7B on 0. The widest gap is MATH, where DeepSeek-V3.2-Exp-Base scores 62.5 against 13.1.

DeepSeek-V3.2-Exp-Base vs Mistral 7B

Across 12 shared benchmarks, DeepSeek-V3.2-Exp-Base scores higher on 12 and Mistral 7B on 0. The widest gap is MATH, where DeepSeek-V3.2-Exp-Base scores 62.5 against 13.1.

DeepSeekvsMistral12 shared benchmarks12–0 head-to-head
BenchmarkDeepSeek-V3.2-Exp-BaseMistral 7B
arc_challenge95.560
bbh88.756.1
C-Eval9147.4
GPQA Diamond5224.7
GSM8K91.152.2
hellaswag89.483.2
humaneval67.729.3
MATH62.513.1
MBPP75.651.1
mmlu87.864.2
MMLU-Pro63.330.9
winogrande85.678.4

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.