DataLearner logo

Kimi K2 Thinking Benchmark Details

Kimi K2 Thinking currently shows benchmark results led by AIME2025 (1 / 106, score 100), Terminal-Bench (4 / 35, score 47.10), HLE (21 / 168, score 51).

Benchmark Results

Kimi K2 Thinking

Benchmark Results

Thinking
Tool usage

General Knowledge

6 evaluations
Benchmark / mode
Score
Rank/total
MMLU Pro
84.60
32 / 126
GPQA Diamond
84.50
56 / 182
LiveBench
Thinking Enabled
61.59
63 / 115
HLE
51
21 / 168
HLE
44.90
40 / 168
HLE
23.90
104 / 168

Coding and Software Engineer

2 evaluations
Benchmark / mode
Score
Rank/total
LiveCodeBench
83.10
22 / 120
SWE-bench Verified
71.30
53 / 110

Math and Reasoning

4 evaluations
Benchmark / mode
Score
Rank/total
AIME2025
100
1 / 106
AIME2025
99.10
12 / 106
AIME2025
94.50
27 / 106
FrontierMath - Tier 4
Thinking Enabled
0
72 / 80

AI Agent - Tool Usage

1 evaluations
Benchmark / mode
Score
Rank/total
Terminal-Bench
47.10
4 / 35

常识推理

1 evaluations
Benchmark / mode
Score
Rank/total
Simple Bench
Standard Mode
39.60
41 / 63

Agent Level Benchmark

1 evaluations
Benchmark / mode
Score
Rank/total

Instruction Following

1 evaluations
Benchmark / mode
Score
Rank/total
IF Bench
68
18 / 30

AI Agent - Information Search

1 evaluations
Benchmark / mode
Score
Rank/total
BrowseComp
60.20
35 / 51

Claw-style Agent Evaluation

1 evaluations
Benchmark / mode
Score
Rank/total
Claw Bench
Thinking EnabledTools
82.50
17 / 29
Compare with other models