DeepSeek-V3.1 Terminus Benchmark Details

DeepSeek-V3.1 Terminus currently shows benchmark results led by SimpleQA (2 / 45, score 96.80), MMLU Pro (17 / 116, score 85), LiveCodeBench (21 / 109, score 80).

Benchmark Results

DeepSeek-V3.1 Terminus

Benchmark Results

综合评估

4 evaluations

Benchmark / mode

Score

Rank/total

MMLU Pro

thinking

17 / 116

GPQA Diamond

thinking

71 / 166

LiveBench

thinking

71.40

15 / 51

HLE

thinking

15.20

92 / 128

编程与软件工程

1 evaluations

Benchmark / mode

Score

Rank/total

LiveCodeBench

thinking

21 / 109

数学推理

1 evaluations

Benchmark / mode

Score

Rank/total

AIME2025

thinking

38 / 107

Compare with other models

DeepSeek-V3.1 Terminus Benchmark Details

DeepSeek-V3.1 Terminus currently shows benchmark results led by SimpleQA (2 / 45, score 96.80), MMLU Pro (17 / 116, score 85), LiveCodeBench (21 / 109, score 80).