Microsoft just released the information that GPT-4.5 preview model is available for deployment at Azure AI Foundry! Not everyone can yet use this, but it requires a registration before it can be used. I, of course, placed in my application for the model but at this point I am still waiting for the registration result. Unfortunately I can’t give my experiences about the model capabilities yet 🔒 – but will do that later once I get the lock removed!
Registration is required to use this model: Access will be granted according to Microsoft’s eligibility criteria; you may not be eligible for access at this time. If you have already requested access, thank you for your patience while your submission is reviewed. Request Access here: https://aka.ms/oai/gptaccess
Microsoft posted an article about the latest updates to Azure AI Foundry: Announcing new models, customization tools, and enterprise agent upgrades in Azure AI Foundry.
Azure AI Foundry description for GPT-4.5:Core capabilities of GPT-4.5Model Qualities: more natural interaction experienceAccuracy & Hallucinations: more precise and relevantStronger human alignment: understanding of nuance, and natural conversationCreative and practical applications: for coding and project managementAvailability and details
Azure AI Foundry description for GPT-4.5:
GPT-4.5 Preview is the largest and strongest general purpose model, best suited for chat for its advancement in delivering more natural chat interactions. It is designed to be highly capable in language tasks, and excels in areas such as intuition, writing style, and knowledge. GPT-4.5 Preview handles creative prompts well and responds effectively to obscure knowledge queries. It has leading scores on the Simple QA benchmark which tests factual knowledge.
GPT-4.5 Preview is a classic pretrained model like GPT-4 and gpt-4o. Compared to previous GPT models, GPT-4.5 Preview is larger and trained for longer. Consider GPT-4.5 Preview as a complementary model to reasoning-focused models such as o1 and o3-mini.

Core capabilities of GPT-4.5
Developers can leverage GPT-4.5 in numerous ways to enhance productivity and creativity. In communication, users can rely on GPT-4.5 to craft clear and effective emails, messages, and documentation. It also offers personalized learning and coaching experiences, helping users to acquire new skills or deepen knowledge in specific areas. During brainstorming sessions, GPT-4.5 can generate innovative ideas and solutions, making it a valuable tool for creative thinking.
For project planning and execution, GPT-4.5 assists users in organizing their tasks, ensuring thorough and efficient approaches. It can also handle complex task automation, simplifying intricate processes and workflows. Developers can streamline their coding workflows by getting step-by-step guidance and automating repetitive tasks, to save time and reduce errors. Overall, GPT-4.5 is a versatile model. Starting today, enterprise customers can access GPT-4.5 in Azure AI Foundry.
Introducing GPT-4.5 in preview on Azure OpenAI Service
Model Qualities: more natural interaction experience
Early testing (not yet by me personally) suggests that interacting with GPT-4.5 Preview feels more natural. It has a broader knowledge base, and should have improved ability to follow user intent, and greater “EQ” to make it useful for tasks like improving writing, programming, and solving practical problems. Microsoft and OpenAI expects it to hallucinate less than other models.
Accuracy & Hallucinations: more precise and relevant
Microsoft’s and OpenAIs testing shows that GPT-4.5 Preview achieves lower hallucination rates compared to GPT-4o (37% vs. GPT-4o at 61.2%) and higher accuracy (61.9% compared to 38.4% for GPT-4o). This means “compared to GPT-4o, developers can rely on more precise and relevant responses”
Stronger human alignment: understanding of nuance, and natural conversation
For GPT-4.5 Preview, Microsoft developed new, scalable alignment techniques that enable training larger and more powerful models with data derived from smaller models. These techniques improve GPT 4.5 Preview’s steerability, understanding of nuance, and natural conversation.
Creative and practical applications: for coding and project management
GPT-4.5 Preview should then excel in writing help, design, multi-step coding workflows, task automation, communication, learning, coaching, and brainstorming. Also shows strong performance in planning, execution, and complex task automation. “making it a more effective tool for coding and project management.“
Availability and details
GPT-4.5 is available currently at East US 2 (Global Standard) and Sweden Central (Global Standard) regions & deployments. The context window size for GPT-4.5 is 128k and it can produce up to 16k tokens on output. Training data is up to October 2023.
Satya Nadella mentioned in LinkedIn about the 4.5 preview model: “GPT-4.5 is now in preview, demonstrating a big step forward in both pre-training and post-training scale. “

What this “post-training” really will mean in the practice I am sure we know very soon. This story will continue, once I get the lock open 🔓
If you want to get your hands on GPT-4.5 then set up an Azure Subscription with Azure AI Foundry and register your scenario to Microsoft for eligibility check.
Registration is required to use this model: Access will be granted according to Microsoft’s eligibility criteria; you may not be eligible for access at this time. If you have already requested access, thank you for your patience while your submission is reviewed. Request Access here: https://aka.ms/oai/gptaccess
Published by
I work, blog and speak about Future Work : AI, Microsoft 365, Copilot, Microsoft Mesh, Metaverse, and other services & platforms in the cloud connecting digital and physical and people together.
I have about 30 years of experience in IT business on multiple industries, domains, and roles.
View all posts by Vesa Nopanen