AI Benchmark Model

DeepSeek V3.1 (Non-reasoning)

Benchmark scores, pricing, speed, and model comparisons for DeepSeek V3.1 (Non-reasoning).

Key scores

Intelligence28.1Index score
Coding28.4Index score
Math49.7Index score
Blended Cost$0.83Per 1M tokens

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

Benchmark results

10 tracked benchmark rows for DeepSeek V3.1 (Non-reasoning).

MMLU Pro Knowledge
83.3%
GPQA Reasoning
73.5%
LiveCodeBench Code
57.7%
AIME 2025 Math
49.7%
LCR Other
45.0%
IFBench Other
37.8%
SciCode Other
36.7%
TAU2 Other
34.8%
TerminalBench Hard Other
24.2%
HLE Other
6.3%

Comments (0)

No comments yet

Be the first to share your thoughts!