Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Threat Intelligence reasoning on CyberSocEval
Loading...
54.25
Accuracy
LLaMa-4-Maverick
4.018
17.059
30.1
43.141
Oct 15, 2025
Accuracy
Jaccard Similarity
Updated 23d ago
Evaluation Results
Method
Method
Links
Accuracy
Jaccard Similarity
LLaMa-4-Maverick
2025.10
54.25
72.57
GPT-4o
2025.10
53.57
73.19
LLaMa-4-Scout
2025.10
50.34
69.9
CyberPal-2.0
Parameters=14B
2025.10
45.07
67.78
Qwen3
Parameters=14B
2025.10
43.54
63.12
Qwen3
Parameters=8B
2025.10
42.86
58.59
CyberPal-2.0
Parameters=8B
2025.10
38.61
65.6
CyberPal-2.0
Parameters=4B
2025.10
19.39
51.85
Qwen3
Parameters=4B
2025.10
5.95
10.12
Feedback
Search any
task
Search any
task