AI Benchmark Model

DeepSeek V3.1 Terminus (Reasoning)

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

Key scores

Intelligence33.9Index score
Coding33.7Index score
Math89.7Index score
Blended Cost$1.91Per 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 (Reasoning).

AIME 2025 Math
89.7%
MMLU Pro Knowledge
85.1%
LiveCodeBench Code
79.8%
GPQA Reasoning
79.2%
LCR Other
65.0%
IFBench Other
57.0%
SciCode Other
40.6%
TAU2 Other
37.1%
TerminalBench Hard Other
30.3%
HLE Other
15.2%

Comments (0)

No comments yet

Be the first to share your thoughts!