360's Tulongfeng achieves a 99.4% success rate and tops the CyberGym vulnerability reproduction evaluation
360 Group's self-developed AI security attack and defense system Tulongfeng topped the CyberGym benchmark test with a 99.4% vulnerability reproduction success rate, competing on the same leaderboard with models and agents from vendors including OpenAI, Anthropic, and Google. The evaluation covers 1,507 tasks, and Tulongfeng's success rates across three independent evaluation rounds were 99.27%, 99.47%, and 99.40%, averaging 99.38%. 360 states that Tulongfeng has discovered more than 15,000 vulnerabilities in total, and more than 500 enterprises have already integrated the system to conduct security testing.
Why read this
AI score 62
360's Tulongfeng has provided three rounds of independent evaluation data on the CyberGym vulnerability reproduction task, which can be compared with the performance of OpenAI, Anthropic, and Google on the same leaderboard.
Topics
Body · AI translation
Related topics top start Related topics top end Futures promotion begin Futures promotion end Miaopai begin Miaopai end Video player start Video player end Market chart begin Market chart end
Sina Tech News, October 30 afternoon news: Recently, in CyberGym, an internationally authoritative benchmark test known as the "World Cup" of AI security, 360 Group's self-developed AI security attack and defense system Tulongfeng topped the leaderboard with a 99.4% vulnerability reproduction success rate and set a new world record. Competing on the same leaderboard were models and agents from global leading AI vendors including OpenAI, Anthropic, and Google.
It is reported that CyberGym focuses on complex real-world challenges, covering a large number of mainstream open-source software and highly difficult historical vulnerabilities, and strictly tests the autonomous reasoning, vulnerability reproduction, and PoC verification capabilities of large models and security agents in environments with massive source code, complex contexts, and multi-stage adversarial conditions. Because it is close to real industrial protection scenarios and has stringent verification standards, it is regarded as one of the most valuable benchmarks globally for evaluating the practical capabilities of code security large models.
Evaluation data shows that across all 1,507 tasks, 360 Tulongfeng underwent three rounds of independent evaluation and verification, with success rates of 99.27%, 99.47%, and 99.40%, respectively, for an average success rate of 99.38%, equivalent to successfully tackling nearly 1,500 real vulnerabilities and setting a new record for the highest score on the leaderboard.
As of now, Tulongfeng has cumulatively discovered more than 15,000 security vulnerabilities, covering multiple fields including operating systems, office software, and AI applications. At the same time, Tulongfeng has passed the China Academy of Information and Communications Technology's capability evaluation for vulnerability mining agents, and more than 500 enterprises have connected to the system to carry out specialized security testing, covering key industries such as finance, energy, government affairs, and manufacturing. news_keyword_pub,stock,sh600628 iframe below main text iframe below main text General statement General statement QR code at end start QR code at end start
QR code at end start QR code at end start Editor name and work code
Editor name and work code