The tempo of AI innovation continues to speed up. Only a few months after the discharge of GPT-5.2 and GPT-5.3, Microsoft Foundry now brings us GPT-5.4—OpenAI’s most succesful frontier mannequin so far to Microsoft Foundry. This launch marks a big leap ahead in reasoning, agentic workflows, and professional-grade automation.
Whether or not you’re constructing AI brokers, automating complicated workflows, or exploring new frontiers in software program improvement and information evaluation, GPT-5.4 is designed to ship sooner, extra dependable, and extra clever outcomes.
What’s in GPT-5.4?
GPT-5.4 introduces a spread of enhancements that elevate its efficiency throughout skilled and enterprise situations:
- Constructed-in agentic workflows for planning and execution
- Native laptop use capabilities (keyboard, mouse, screenshots)
- Device Seek for navigating massive instrument ecosystems
- Help for very large context windows (up to 1,050,000 tokens)
- Improved token effectivity for sooner, lower-cost responses
- Enhanced coding and software program automation reliability
- Larger factual accuracy and decreased hallucinations
These capabilities make GPT-5.4 a strong instrument for doc and spreadsheet creation, coding, information evaluation, and long-form reasoning duties.
Introducing GPT-5.4 Professional
For situations the place analytical depth and completeness are extra essential than latency, Microsoft Foundry additionally provides GPT-5.4 Professional—a premium variant of the mannequin.
GPT-5.4 Professional is designed for deep analytical workflows, corresponding to scientific analysis, strategic decision-making, and sophisticated problem-solving. It introduces:
- Multi-path reasoning analysis to discover various options
- Higher analytical depth for issues with trade-offs or a number of legitimate outcomes
- Improved stability throughout lengthy reasoning chains
- Enhanced choice assist the place rigor outweighs pace
With a bigger context window (400,000 tokens for the time being, bigger context window coming quickly) and the identical excessive output capability (128,000 tokens), GPT-5.4 Professional is the go-to mannequin for high-assurance, high-complexity duties.
Capabilities
That can assist you consider the evolution of those fashions, right here’s a side-by-side comparability of GPT-5.4, GPT-5.4 Professional, and GPT-5.2:
| Functionality | GPT-5.4 | GPT-5.4 Professional | GPT-5.2 |
|---|---|---|---|
| Reasoning | Stronger reasoning for complicated, multi-step duties | Multi-path reasoning analysis for deeper evaluation | Adaptive reasoning for complicated queries |
| Agentic Workflows | Constructed-in agentic workflows for planning and execution | Enhanced agentic workflows with improved stability | Accelerates agent improvement |
| Pc Interplay | Native laptop use (keyboard, mouse, screenshots) | Identical as GPT-5.4 | Not obtainable |
| Device Administration | Device Seek for massive instrument ecosystems | Identical as GPT-5.4 | Dependable instrument use and ruled integrations |
| Token Effectivity | Improved for sooner, lower-cost responses | Identical as GPT-5.4 | Improved over GPT-5.1 |
| Coding Reliability | Enhanced software program automation and code era | Identical as GPT-5.4 | Dependable code era and modernization |
| Factual Accuracy | Larger factual accuracy and decreased hallucinations | Identical as GPT-5.4 | Higher consistency and accuracy |
| Context Window | 1,050,000 tokens | 400,000 tokens (1,050,000 coming quickly) | 400,000 tokens |
| Context Reminiscence – Output | 128,000 tokens | 128,000 tokens | 128,000 tokens |
| Coaching Information Cutoff | August 2025 | August 2025 | August 2025 |
| Finest For | Dependable execution, agentic workflows, software program automation | Scientific analysis, complicated decision-making, deep analytical workflows | Lengthy-form reasoning, structured content material, enterprise brokers |
| Latency vs. Depth | Prioritizes pace and reliability | Prioritizes analytical depth and completeness over latency | Balanced efficiency |
Pricing and Availability
Microsoft Foundry provides versatile pricing for GPT-5.4 primarily based on context size:
- GPT‑5.4 (<272K enter tokens):
- $2.50 per million enter tokens
- $0.25 per million cached enter tokens
- $15.00 per million output tokens
- GPT‑5.4 (>272K enter tokens):
- $5.00 per million enter tokens
- $0.50 per million cached enter tokens
- $22.50 per million output tokens
- GPT‑5.4 Professional:
- $30.00 per million enter tokens
- $180.00 per million output tokens
At launch, GPT-5.4 is obtainable in Customary World and Customary Information Zone (US), whereas GPT-5.4 Professional is obtainable in Customary World, with extra deployment choices anticipated quickly.
Ideas
GPT-5.4 is a superb leap ahead in enterprise AI. With its large context window, built-in agentic capabilities, and native laptop interplay, it’s constructed for the subsequent era {of professional} work. And for many who want much more analytical depth, GPT-5.4 Professional provides unmatched reasoning energy.
The way forward for AI within the enterprise is arriving sooner than ever—and Microsoft Foundry is making it simpler to undertake and scale these capabilities securely and effectively.
What are you most excited to construct with GPT-5.4?
Assets
Printed by
I work, weblog and discuss Future Work : AI, Microsoft 365, Copilot, Loop, Azure, and different providers & platforms within the cloud connecting digital and bodily and folks collectively.
I’ve 30 years of expertise in IT enterprise on a number of industries, domains, and roles.
View all posts by Vesa Nopanen
You might also like
More from Metaverse Global
Logitech MX Creative Console Review: The Ultimate Editing Tool?
For years, I’ve relied solely on a posh net of keyboard shortcuts to muscle my approach by means of …
GCOIN Debuts as Playnance Accelerates Ecosystem Growth
The evolution of Web3 leisure is being …
