AI Benchmark Model

DeepSeek V3.2 Exp (Reasoning)

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

Key scores

Intelligence32.9Index score
Coding33.3Index score
Math87.7Index score
Blended Cost$0.31Per 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 Exp (Reasoning).

AIME 2025 Math
87.7%
MMLU Pro Knowledge
85.0%
GPQA Reasoning
79.7%
LiveCodeBench Code
78.9%
LCR Other
69.0%
IFBench Other
54.1%
SciCode Other
37.7%
TAU2 Other
33.9%
TerminalBench Hard Other
31.1%
HLE Other
13.8%

Comments (0)

No comments yet

Be the first to share your thoughts!