Local Developer Alican Kiraz’s Artificial Intelligence Model Made Our Country Proud
The Titus-CybersecurityLLM model developed by our local developer Alican Kiraz has achieved global success on the CS-Eval platform.
Developer Alican Kiraz has achieved remarkable success with the new artificial intelligence model he customized in the field of cyber security. The model named Titus-CybersecurityLLM-36B-A3B-v1, developed based on Qwen3.6-35B, has occupied a valuable position in the internationally accepted CS-Eval platform.
CS-Eval, prepared by Fudan University and the University of Chinese Academy of Science, is known as one of the most comprehensive evaluation platforms focused on cyber security. The model developed by Alican Kiraz ranked 9th worldwide among both private and public use models on this platform.
Significant Success in Global Rankings
The Titus-CybersecurityLLM-36B-A3B-v1 model stands out with its comprehensive average of 92.42 points. This work, designed with the MoE architecture, has been optimized to demonstrate high performance, especially in cyber security tasks.
The model’s performance data was measured in detail in critical areas such as system security, access control, encryption technologies and threat detection. According to the results obtained, the model managed to rank first in the world in the open source models category.
Detailed Performance Analysis
The information shared on the CS-Eval platform reveals the competence of the model in different security disciplines. Titus-CybersecurityLLM-36B-A3B-v1 received 94.67 points in basic system and software security and 95.02 points in supply chain security.
The model, which achieved 93.73 points in the threat detection and prevention heading, also showed a stable performance in issues such as infrastructure security and data confidentiality. The model, which was included in the system on June 24, 2026, has gained a solid place among cybersecurity-focused artificial intelligence solutions.
The success of the model in question once again reveals the potential of customized artificial intelligence models in the field of cyber security. This study, developed by Alican Kiraz, is expected to contribute to cyber security processes.
What do you think about this development?
