AI Squawk the real-time terminal for AI news

↗

TOOLING CWE-bench v2: More coverage, quality, and transparency

10d ago · Hugging Face Blog →

CWE-bench is designed to test the limits of frontier cyber capabilities.

Why is this ranked here?

Source tier 1: +40
Category TOOLING: +15
Base score: 55
Velocity (0 corroborating sources): +0
Age decay (10d since published): −55
Effective score right now: 0