AI Benchmark Model

DeepSeek V3.1 Terminus (Non-reasoning)

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

Key scores

Intelligence28.5Index score
Coding31.9Index score
Math53.7Index score
Blended Cost$0.45Per 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 Terminus (Non-reasoning).

MMLU Pro Knowledge
83.6%
GPQA Reasoning
75.1%
AIME 2025 Math
53.7%
LiveCodeBench Code
52.9%
LCR Other
43.3%
IFBench Other
41.2%
TAU2 Other
37.1%
SciCode Other
32.1%
TerminalBench Hard Other
31.8%
HLE Other
8.4%

Comments (0)

No comments yet

Be the first to share your thoughts!