OpenAI has launched GPT-5.1-Codex-Max, an upgraded coding model with improved performance metrics over its predecessor. It excels in various software engineering tasks but still faces challenges in cybersecurity capabilities. The article critiques the model's evaluations and compares it to previous versions, raising questions about its real-world usefulness.

OpenAI has released GPT-5.2-Codex, an advanced coding model designed for software development and cybersecurity. It enhances long-context understanding, tool reliability, and cybersecurity capabilities, enabling more effective coding and threat detection. The release aims to balance accessibility with safety in deployment.