AI Benchmark Model

DeepSeek V3.2 (Reasoning)

Benchmark scores, pricing, speed, and model comparisons for DeepSeek V3.2 (Reasoning).

Key scores

Intelligence41.7Index score
Coding36.7Index score
Math92.0Index score
Blended Cost$0.34Per 1M tokens

Review benchmark scores, pricing, performance data, and generated comparisons for this AI model.

Benchmark results

10 tracked benchmark rows for DeepSeek V3.2 (Reasoning).

AIME 2025 Math
92.0%
TAU2 Other
90.6%
MMLU Pro Knowledge
86.2%
LiveCodeBench Code
86.2%
GPQA Reasoning
84.0%
LCR Other
65.0%
IFBench Other
60.7%
SciCode Other
38.9%
TerminalBench Hard Other
35.6%
HLE Other
22.2%

Comments (0)

No comments yet

Be the first to share your thoughts!