OpenAI has introduced two compact models, GPT-5.4 Mini and GPT-5.4 Nano, with the goal of diminishing the performance gap with their flagship counterparts while offering lower latency and cost efficiency. Notably, the GPT-5.4 Mini outperforms its predecessors in crucial domains, such as programming, boasting operation speeds over twice as fast. It is now accessible through the OpenAI API, the Codex platform, and ChatGPT. On the other hand, GPT-5.4 Nano is tailored as a cost-effective, low-latency lightweight solution, exclusively available to developers via the API. Assessments indicate that the Mini variant excels in programming and multimodal tasks, nearing the performance benchmarks set by flagship models. The Nano, while showcasing marked enhancements over its forerunners in certain tasks, exhibits slightly subpar performance in visual tasks. OpenAI underscores the significance of these two new models within a multi-model tiered framework, where collaborative efforts among multiple models have emerged as a novel paradigm in product design, aiding developers in curbing inference expenses. Concurrently, OpenAI warns of potential risks in the market and advises investors to proceed with caution.