Microsoft unveils cheaper, faster AI image model

Microsoft has launched MAI-Image-2-Efficient, a more affordable and speedier version of its text-to-image AI model. This new variant aims to provide production-ready quality at nearly half the cost of its predecessor, making it suitable for high-volume, cost-sensitive workloads like product photography and marketing creative. The model also boasts significant speed and efficiency improvements, outperforming competing hyperscaler models in latency benchmarks. This release signals Microsoft's strategic move to build a more self-sufficient AI stack, reducing reliance on partners like OpenAI and optimizing costs for future AI agent applications.