OpenAI’s latest AI models are designed for speed


Latest models for ChatGPT Users and developers using OpenAI’s API are designed to be a workforce, designed for tasks like critical programming, where large, powerful AI models are expensive. GPT-5.4 Mini and Nano, which OpenAI was released on Tuesdayare the smallest and fastest versions of GPT-5.4, which It debuted earlier this month.

Atlas of Artificial Intelligence

These new models are part of OpenAI’s efforts to go programmatic as it competes with rival Anthropic in the AI ​​software engineering market. Latest OpenAI models for Codex Alimentarius Codecs have posed a direct challenge to anthropology Claude Codeany It went viral At the end of 2025 due to its ability to do so Create applications from scratch.

Disclosure: Ziff Davis, the parent company of CNET, in 2025 filed a lawsuit against OpenAI, alleging that it infringed Ziff Davis’s copyrights in training and operating its AI systems.

OpenAI says GPT-5.4 mini is twice as fast as its predecessor, GPT-5 mini, at tasks like programming, reasoning, and tool use. In some benchmarks, it is close to the capabilities of standard GPT-5.4.

Read more: Best Smart Chatbots of 2026

The company suggested that GPT-5.4 mini is best for things like code editing and debugging. It can be used as a subproxy in Codex, with a larger model such as GPT-5.4 delegating certain tasks to the faster, cheaper model.

GPT-5.4 nano is smaller, and was proposed by OpenAI for heavy-duty work like data classification and extraction.

GPT-5.4 mini will be available to developers through the API and through Codex and ChatGPT. ChatGPT Free and Go users can access it through the “Reflection” feature. Others will find it as a fallback model when they reach the GPT-5.4 think rate limit. GPT-5.4 nano is only available in the Application Programming Interface (API).



Leave a Reply

Your email address will not be published. Required fields are marked *