AI Benchmark Model

DeepSeek V3.2 (Non-reasoning)

Benchmark scores, pricing, speed, and model comparisons for DeepSeek V3.2 (Non-reasoning).

Key scores

Intelligence32.1Index score
Coding34.6Index score
Math59.0Index score
Blended Cost$0.78Per 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 (Non-reasoning).

MMLU Pro Knowledge
83.7%
TAU2 Other
78.9%
GPQA Reasoning
75.1%
LiveCodeBench Code
59.3%
AIME 2025 Math
59.0%
IFBench Other
49.0%
LCR Other
39.0%
SciCode Other
38.7%
TerminalBench Hard Other
32.6%
HLE Other
10.5%

Comments (0)

No comments yet

Be the first to share your thoughts!