OpenAI has announced the release of GPT‑5.4, the latest advancement in its series of large language models, offering enhanced performance and capabilities specifically tailored for professional and enterprise use. The rollout began on 5 March 2026 across ChatGPT, the OpenAI API, and the Codex developer platform.
According to OpenAI’s press release, GPT‑5.4 is positioned as the company’s most capable and efficient “frontier model” to date, designed to handle complex real‑world workflows with improved accuracy, reasoning, and tool integration. The model introduces two primary configurations — GPT‑5.4 Thinking for advanced reasoning tasks and GPT‑5.4 Pro for high‑performance, enterprise‑scale workloads.
Enhanced Capabilities for Knowledge Work
OpenAI says GPT‑5.4 consolidates recent progress in reasoning, coding and agentic workflows into a single foundational model. The new release incorporates and builds on the coding strengths of the prior GPT‑5.3‑Codex model, while improving performance across professional tasks such as generating documents, managing spreadsheets and creating presentations.
A distinctive feature of GPT‑5.4 is its ability to provide an upfront plan for multi‑step problem‑solving within ChatGPT, allowing users to adjust direction mid‑task based on evolving requirements. This “steerability” is expected to reduce iteration cycles and enhance productivity for users working on complex assignments.
New Technical Benchmarks
GPT‑5.4 supports a context window of up to 1 million tokens, enabling the model to process significantly larger text inputs and workflows in a single interaction — a capability that considerably expands its usefulness for long‑form tasks such as deep research, report drafting and comprehensive code development.
Early benchmark results show that GPT‑5.4 achieves state‑of‑the‑art performance in professional benchmarks, including an 83 % score on the GDPval evaluation, which measures knowledge work across diverse occupations. This marks a notable improvement over previous iterations and suggests the model can match or exceed professional outputs in many domains.
Computer‑Use and Tool Integration
Another major advancement is GPT‑5.4’s native computer‑use capabilities, enabling it to perform tasks across software environments — including issuing keyboard and mouse commands based on visual inputs — a first for general‑purpose models from OpenAI. This functionality promises new automation possibilities for agents and developers alike.
The model also introduces tool-search mechanisms to streamline interactions with large ecosystems of external tools and connectors, reducing overhead and improving token efficiency in complex workflows.
Reduced Errors and Expanded Availability
OpenAI highlights significant improvements in factual reliability, noting that GPT‑5.4 produces fewer incorrect claims and fewer overall errors than its predecessors. The company also emphasises continued attention to safety monitoring and evaluation as capabilities grow.
GPT‑5.4 is already available to ChatGPT Plus, Team and Pro subscribers, as well as through the OpenAI API and Codex platforms. Pricing varies by use case and tier, with the Pro tier intended for organisations that require maximum performance for large‑scale or mission‑critical tasks.
Director
Bio: An (HND, BA, MBA, MSc) is a tech-savvy digital marketing professional, writing on artificial intelligence, digital tools, and emerging technologies. He holds an HND in Marketing, is a Chartered Marketer, earned an MBA in Marketing Management from LAUTECH, a BA in Marketing Management and Web Technologies from York St John University, and an MSc in Social Business and Marketing Management from the University of Salford, Manchester.
He has professional experience across sales, hospitality, healthcare, digital marketing, and business development, and has worked with Sheraton Hotels, A24 Group, and Kendal Nutricare. A skilled editor and web designer, He focuses on simplifying complex technologies and highlighting AI-driven opportunities for businesses and professionals.