GPT-5.4
OpenAI's most capable frontier model unifying reasoning, coding, and agentic computer use in a single system
GPT-5.4
OpenAI • March 2026
Training Data
Up to August 2025
GPT-5.4
March 2026
Parameters
Not disclosed
Training Method
Unified Agentic Architecture
Context Window
1,050,000 tokens
Knowledge Cutoff
August 2025
Key Features
Native Computer Use • 1M Token Context • Token-Efficient Reasoning • Agentic Workflows
Capabilities
Professional Tasks: Outstanding (83% GDPval)
Accuracy: Exceptional (33% fewer errors)
Coding: Maximum
What's New in This Version
33% fewer false claims than GPT-5.2, native computer control, and consolidates GPT-5.3-Codex coding into a unified model
OpenAI's most capable frontier model unifying reasoning, coding, and agentic computer use in a single system
What's New in This Version
33% fewer false claims than GPT-5.2, native computer control, and consolidates GPT-5.3-Codex coding into a unified model
Technical Specifications
Key Features
Capabilities
Other OpenAI Models
Explore more models from OpenAI
GPT-6.1 Sol
An upgrade to GPT-6 Sol that nearly matches GPT-6 Astra on agentic coding, computer use, and professional work at one-fifth of Astra's token prices
GPT-6 Sol
The GPT-6 workhorse for complex coding and agentic workflows — Astra-level reliability at a fraction of Astra's price
GPT-6 Luna
OpenAI's most efficient model for focused, high-volume tasks — summarizing documents, extracting information, and answering quick questions
