OpenAI Introduces GPT‑5.4 Mini and Nano

OpenAI has unveiled GPT‑5.4 Mini and Nano, their most advanced compact models to date.

These models inherit many of GPT‑5.4's strengths while being optimised for speed and efficiency, making them ideal for high-volume workloads.

The GPT‑5.4 Mini model offers significant improvements over its predecessor, GPT‑5 Mini, particularly in coding, reasoning, multimodal understanding, and tool utilisation. It operates more than twice as fast and approaches the performance of the larger GPT‑5.4 model in several evaluations, such as SWE-Bench Pro and OSWorld-Verified.

On the other hand, GPT‑5.4 Nano is the smallest and most cost-effective version of GPT‑5.4, designed for tasks prioritising speed and cost. It represents a substantial upgrade over GPT‑5 Nano, recommended for classification, data extraction, ranking, and coding subagents handling simpler tasks.

---

Want to hear more? Join Mal & Matt on the Property AI Report Podcast each week!

Access from your preferred podcast provider by clicking here

---

These models are tailored for environments where latency significantly impacts the user experience. They are particularly suited for coding assistants requiring responsiveness, subagents completing tasks swiftly, systems interpreting screenshots, and multimodal applications reasoning over images in real-time. In such scenarios, the optimal model is often not the largest but the one that can respond quickly and perform well on complex tasks.

In coding workflows, GPT‑5.4 Mini and Nano excel in fast iteration, handling targeted edits, codebase navigation, front-end generation, and debugging with low latency. This makes them ideal for coding tasks demanding speed and cost-efficiency.

GPT‑5.4 Mini also shines in systems combining models of different sizes. In Codex, for instance, a larger model like GPT‑5.4 manages planning and coordination, while GPT‑5.4 Mini subagents tackle narrower subtasks in parallel. This approach becomes more beneficial as smaller models become faster and more capable.

For computer use, GPT‑5.4 Mini excels in multimodal tasks, quickly interpreting dense user interfaces to complete tasks efficiently. It approaches GPT‑5.4's performance on OSWorld-Verified, surpassing GPT‑5 Mini significantly.

Availability-wise, GPT‑5.4 Mini is accessible in the API, Codex, and ChatGPT, supporting a wide range of functionalities. GPT‑5.4 Nano is available in the API, offering a cost-effective solution for specific tasks.

For further details on model safeguards, users can refer to the System Card addendum on OpenAI's Deployment Safety Hub.

---

Want to hear more? Join Mal & Matt on the Property AI Report Podcast each week!

Access from your preferred podcast provider by clicking here

---

Made with TRUST_AI - see the Charter: https://www.modelprop.co.uk/trust-ai