#GPT-5

In a benchmark test simulating social dynamics, seven LLMs played the game Werewolf. GPT-5 significantly outperformed the others with a 96.7% win rate, demonstrating superior strategic thinking and manipulation skills. Other models, including Qwen3 and Kimi-K2, showed respectable performance. Analysis revealed distinct personality traits in each model; for example, Kimi-K2 exhibited aggressive tactics. The experiment highlights the importance of social skills for AI agents operating within human teams, alongside traditional benchmarks.

GPT-5, OpenAI’s latest AI model, launched on August 8th, is generating significant attention. CEO Sam Altman compared its predecessors to students: GPT-3 as a bright but uneven high schooler, and GPT-4 as a college student. The specifics of GPT-5’s enhanced capabilities remain to be fully explored.

OpenAI has launched GPT-5, touted as a significant upgrade with expert-level performance in programming, mathematics, writing, healthcare, and visual perception. GPT-5 features an intelligent model, a deep reasoning engine, and a real-time router. Free and paid users will get access, with subscribers enjoying higher usage. GPT-5 demonstrates improvements in response time, accuracy, instruction following, and reduced hallucination. It excels in programming, creative writing, and healthcare. GPT-5 Pro surpasses o3-Pro for complex tasks, achieving state-of-the-art results in GPQA.

Baidu is expected to launch its latest Ernie inference model by late August, reportedly surpassing OpenAI’s o3 in reasoning and computation. An updated Ernie foundational model is also planned. This occurs as OpenAI prepares to launch GPT-5, a unified, multi-modal system. Enhanced inference capabilities are becoming vital amid increasing competition in the AI landscape. Baidu, the first Chinese company with a ChatGPT-like chatbot, aims to compete with OpenAI and others by upgrading its models.

OpenAI is teasing a livestream event on X, sparking speculation about the unveiling of GPT-5. A typo in the announcement fueled the rumors. CEO Sam Altman hinted at GPT-5’s integration of several technologies. The standard ChatGPT version may offer free, limited access to GPT-5’s basic capabilities. Plus and Pro subscribers could gain access to more advanced features like voice integration, enhanced search, and research tools.