DataLearner logo

DeepSeek V3.2-Exp Benchmark Details

DeepSeek V3.2-Exp currently shows benchmark results led by SimpleQA (1 / 45, score 97.10), Aider-Polyglot (11 / 59, score 74.20), MMLU Pro (25 / 126, score 85).

Benchmark Results

DeepSeek V3.2-Exp

Benchmark Results

Thinking

General Knowledge

9 evaluations
Benchmark / mode
Score
Rank/total
MMLU Pro
85
25 / 126
MMLU Pro
84
37 / 126
GPQA Diamond
79.90
78 / 179
GPQA Diamond
74
97 / 179
LiveBench
Standard Mode
49.85
91 / 115
LiveBench
Thinking Mode
58.90
73 / 115
HLE
20.30
104 / 159
HLE
19.80
106 / 159
HLE
8.60
139 / 159

Common Sense

1 evaluations
Benchmark / mode
Score
Rank/total
SimpleQA
97.10
1 / 45

Coding and Software Engineer

3 evaluations
Benchmark / mode
Score
Rank/total
LiveCodeBench
74.10
41 / 120
LiveCodeBench
55
84 / 120
SWE-bench Verified
67.80
67 / 108

Math and Reasoning

2 evaluations
Benchmark / mode
Score
Rank/total
AIME2025
89.30
39 / 106
AIME2025
58
83 / 106

AI Agent - Tool Usage

2 evaluations
Benchmark / mode
Score
Rank/total
Terminal-Bench
37.70
14 / 35
Terminal-Bench
23
30 / 35

Agent Level Benchmark

5 evaluations
Benchmark / mode
Score
Rank/total
Aider-Polyglot
Standard Mode
70.20
17 / 59
Aider-Polyglot
Thinking Mode
74.20
11 / 59
τ²-Bench
66.70
26 / 40
τ²-Bench - Telecom
34
34 / 35
τ²-Bench - Telecom
34
34 / 35

Instruction Following

1 evaluations
Benchmark / mode
Score
Rank/total
IF Bench
54.10
26 / 29

AI Agent - Information Search

1 evaluations
Benchmark / mode
Score
Rank/total
BrowseComp
40.10
41 / 45
Compare with other models