AI Hub
← All models

DeepSeek-V3.2 vs DeepSeek V3.2 Speciale

DeepSeek vs DeepSeek — benchmarks, pricing, and capabilities side by side.

  • DeepSeek V3.2 Speciale has the higher intelligence index (89.9 vs 87.1)
  • DeepSeek-V3.2 is cheaper ($0.25 vs $0.29 per 1M input)
  • DeepSeek V3.2 Speciale has a larger context window (164K)
DeepSeek-V3.2DeepSeek V3.2 Speciale
Intelligence index87.189.9
DeveloperDeepSeekDeepSeek
TypeLLMLLM
AccessOpen weightsOpen weights
Context window131,072 tokens163,840 tokens
Input price$0.25 / 1M$0.29 / 1M
Output price$0.38 / 1M$0.43 / 1M
Speed
ReleasedDecember 1, 2025December 1, 2025
Parameters671B (37B active)
Input modalitiesTextText
Output modalitiesTextText

Shared benchmarks

DeepSeek-V3.2
DeepSeek V3.2 Speciale
AIME 2025
92
96.7
GPQA Diamond
84
87.1
Humanity’s Last Exam
22.2
26.1
LiveCodeBench
86.2
89.6
MMLU-Pro
86.2
86.3