The big headlines on this release are efficiency, with OpenAI reporting that GPT-5.4 uses far fewer tokens (47% fewer on some tasks) than its predecessors).

Arguably the biggest new feature of GPT-5.4 is that users will be able to interrupt the model mid-respose if it starts going down the wrong rabbit hole. According to OpenAI, when using GPT-5.4 in ...

The company bills API users based on the number of tokens that its LLMs process while generating a prompt response. A token is a unit of data that comprises a few letters or characters. OpenAI says ...

The GPT 5.4 model brings reasoning, coding, and agentic workflows into a single professional-level AI. It's now available ...

While OpenAI is bringing GPT-5.4 to its API and its AI-powered coding tool, Codex, it’s rolling out its reasoning model, ...

OpenAI debuted its most capable model yet under pressure from a mass user exodus tied to the company's controversial Pentagon ...

OpenAI launches GPT-5.4 across ChatGPT, API, and Codex with stronger reasoning, coding, and computer use capabilities.

When paired with an AI agent system, GPT-5.4 can click a mouse, type keyboard commands, browse the web, and control computer apps.

GPT-5.4 is billed as "our most capable and efficient frontier model for professional work." ...

OpenAI released GPT-5.4 today with native computer use, a 1M-token context window, and new professional benchmarks. Find what's genuinely new.

GPT-5.4 is also more reliable, producing 18% fewer errors and 33% fewer false claims than GPT-5.2, according to OpenAI.

GPT-5.3 Instant reduces hallucinations by 26.8% on web queries and 19.7% on internal knowledge — OpenAI's most-used model now ...