Skip to main content

GPT-5.4

Overview

GPT-5.4 is OpenAI's flagship frontier model released on March 5, 2026, the first mainline model to unify reasoning, coding (GPT-5.3-Codex), and computer use into a single architecture. With up to 1,050,000 tokens of context and 128,000 tokens of max output, it is OpenAI's most capable and versatile model to date.

Key Features

  • Unified Architecture: Merges reasoning, coding, and computer use into one model — no need to switch between specialized models.
  • Configurable Reasoning Effort: Five discrete reasoning levels (none, low, medium, high, xhigh) let developers control thinking depth and cost per query.
  • Computer Use API: A new Computer Use API enables the model to see screens, move cursors, click elements, type text, and interact with desktop applications.
  • Tool Search: Deferred tool loading mechanism that fetches tool definitions only when needed, reducing total token usage by 47% while maintaining the same accuracy.
  • 1M+ Context Window: Supports up to 1,050,000 tokens of context (922K input + 128K output), enabling analysis of entire codebases or document collections in a single request.

Best Use Cases

  • Agentic Coding Assistants: Scores 57.7% on SWE-Bench Pro, ideal for complex multi-step coding tasks and autonomous code repair.
  • Desktop Automation & RPA: OSWorld score of 75% surpasses the human expert baseline of 72.4%, suitable for browser navigation, form filling, and desktop application control.
  • Knowledge-Intensive Work: GDPval score of 83% with 33% fewer factual errors per claim than GPT-5.2, ideal for research analysis, document processing, and professional Q&A.
  • Long-Context Analysis: The 1M token context window is perfect for legal document review, large-scale code audits, and cross-document correlation analysis.

Capabilities and Limitations

CapabilityDetailed Description
Reasoning AbilitySWE-Bench Pro 57.7%, SWE-Bench Verified ~80%, high GPQA Diamond scores, with five configurable reasoning levels.
Creative AbilityExcellent long-form text and code generation with 128K max output supporting whole-project generation.
Multimodal AbilitySupports text and image input with text output; MMMU Pro score of 81.2%.
Response SpeedAs a flagship model, moderate inference speed; higher latency in xhigh reasoning mode, near real-time in none mode.
Context WindowStandard 272K tokens; expandable to 1,050,000 tokens (requires explicit configuration).
Max Output128,000 tokens
Knowledge CutoffAugust 31, 2025

Credits and Pricing

ModelInput (Credits/Token)Output (Credits/Token)
GPT-5.42.5015.00