Microsoft Just Launched 7 AI Models, And It Wants to Challenge OpenAI
Microsoft Just Launched 7 AI Models, And It Wants to Challenge OpenAI
Microsoft is no longer content with being the company that brings other people's AI to the world. It wants to build its own intelligence.
The tech giant has unveiled seven new AI models designed to tackle everything from coding and cybersecurity to image generation, voice conversations, transcription, and decision-making. And the message behind the launch is clear: Microsoft wants to compete more aggressively with AI companies like OpenAI and Anthropic while giving businesses powerful AI tools that cost less to use.
But this is about more than launching new AI models. Microsoft is slowly building a future where it has more control over the AI technology powering its products, businesses, and millions of users.
Meet Microsoft's 7 New AI Models
Rather than building one AI model expected to do everything, Microsoft is developing specialized models designed to handle specific tasks efficiently.
Here is what the new lineup brings:
1. MAI Cyber 2 Flash
This model is designed for cybersecurity. It helps security teams detect vulnerabilities, investigate threats, and respond to security problems faster.
Microsoft says it can deliver performance comparable to GPT-6 Astra on a cybersecurity benchmark at a fraction of the cost. The idea is to make AI-powered security monitoring more affordable, especially for organizations that need to protect their systems around the clock.
2. MAI Code 2
Meet Microsoft's coding assistant built for serious software development.
MAI Code 2 can help developers debug difficult problems, work across multiple files, build software features, understand entire code repositories, and create interfaces.
Microsoft says it approaches the performance of leading rival coding models while costing significantly less to run. It's designed for the kind of everyday coding work businesses and developers actually need to get done.
3. MAI Code 2 Flash
Not every coding task requires a heavyweight AI model.
Code 2 Flash is built for faster, repetitive jobs such as explaining code, making small edits, fixing routine bugs, and helping developers work through quick changes.
It aims to deliver good coding assistance without making users pay for more AI computing power than their tasks require.
4. MAI Image 2.7
Microsoft is also going after the AI image market.
MAI Image 2.7 focuses on image generation and editing, helping users transform existing images and create visuals from prompts.
Microsoft says its image-editing performance is leading compared with other models tested by Artificial Analysis. If those results hold up across real-world use, it could become an interesting alternative for designers, marketers and creators who rely on AI for visual content.
5. MAI Realtime 1
This is Microsoft's push into more natural AI conversations.
MAI Realtime 1 is designed for live speech-to-speech interactions, meaning users can speak to AI and receive spoken responses without the experience feeling like a slow back-and-forth conversation.
The model is built around low latency, making it useful for voice assistants, customer service, conversational AI applications, and businesses building voice-powered experiences.
Microsoft says its pricing starts at just $0.01 per audio minute.
6. MAI Transcribe 2
For anyone who has ever spent hours converting meetings, interviews, or recordings into written text, transcription AI can be a huge time-saver.
MAI Transcribe 2 is built to convert speech into text, with a particular focus on speed and multilingual accuracy.
Microsoft says the technology is already powering Dragon Copilot, which handles millions of healthcare encounters monthly. Better transcription could help professionals spend less time typing notes and more time doing the actual work.
7. Decision 1.1
This one is particularly interesting because it tackles a problem businesses face as AI usage grows: deciding which AI model should handle each task.
Decision 1.1 is designed to help applications classify information, prioritize tasks, verify outputs, and route requests to appropriate AI systems.
Instead of sending every request to the most expensive and powerful model, businesses can use a decision model to determine when a simpler, cheaper model is enough.
That could mean lower AI bills, faster responses, and more efficient AI-powered workflows.
Microsoft Is Making AI Cheaper, And That Is a Big Deal
One of the biggest themes behind Microsoft's announcement is cost.
AI can be incredibly powerful, but businesses using it at scale have to think about the bill. When AI agents are working continuously, analyzing information, writing code, and completing tasks, costs can quickly add up.
Microsoft is positioning its new models as a way to get more AI work done without paying premium prices for every request.
The company claims its MAI Code 2 delivers around 98% of GPT-6.1 Sol's coding performance at roughly 40% of the price, based on its selected coding benchmarks.
That does not automatically mean it will outperform OpenAI in every coding situation. Benchmarks are useful, but real-world performance depends on the task, the software being built, and how well a model handles unexpected problems.
Still, the pricing strategy is important. If Microsoft can deliver reliable results at significantly lower costs, developers and businesses may have a genuine reason to switch or at least start using more than one AI provider.
Is Microsoft Trying to Move Away From OpenAI?
This is where things get even more interesting.
Microsoft and OpenAI have been major partners for years. Microsoft invested heavily in OpenAI and integrated its AI technology into products such as Copilot, Azure, and other services.
But the AI industry is changing fast, and Microsoft has increasingly been developing its own models while also working with other AI companies.
Building its own AI technology gives Microsoft more control over pricing, performance, infrastructure, and the products it wants to develop. It also means Microsoft does not have to depend on a single AI provider for every major AI feature it offers.
This does not mean Microsoft and OpenAI are suddenly enemies or that their partnership has disappeared. Rather, Microsoft is expanding its options and making sure it can compete in a market where the companies supplying AI technology are becoming just as important as the companies building products around it.