Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More<\/em><\/p>\n\n\n\n Microsoft has introduced a new class of highly efficient AI models that process text, images and speech simultaneously while requiring significantly less computing power than other available systems. The new Phi\u20134 models, released today, represent a breakthrough in the development of small language models (SLMs) that deliver capabilities previously reserved for much larger AI systems.<\/p>\n\n\n\n Phi\u20134\u2013multimodal, a model with just 5.6 billion parameters, and Phi-4-Mini, with 3.8 billion parameters, outperform similarly sized competitors and on certain tasks even match or exceed the performance of models twice their size, according to Microsoft\u2019s technical report.<\/p>\n\n\n\n \u201cThese models are designed to empower developers with advanced AI capabilities,\u201d said Weizhu Chen, vice president, generative AI at Microsoft. \u201cPhi-4-multimodal, with its ability to process speech, vision and text simultaneously, opens new possibilities for creating innovative and context-aware applications.\u201d<\/p>\n\n\n\n This technical achievement comes at a time when enterprises are increasingly seeking AI models that can run on standard hardware or at the \u201cedge\u201d \u2014 directly on devices rather than in cloud data centers \u2014 to reduce costs and latency while maintaining data privacy.<\/p>\n\n\n\n What sets Phi-4-multimodal apart is its novel \u201cMixture of LoRAs\u201d technique, enabling it to handle text, images and speech inputs within a single model.<\/p>\n\n\n\n \u201cBy leveraging the Mixture of LoRAs, Phi-4-Multimodal extends multimodal capabilities while minimizing interference between modalities,\u201d the research paper states. \u201cThis approach enables seamless integration and ensures consistent performance across tasks involving text, images, and speech\/audio.\u201d<\/p>\n\n\n\n The innovation allows the model to maintain its strong language capabilities while adding vision and speech recognition without the performance degradation that often occurs when models are adapted for multiple input types.<\/p>\n\n\n\n The model has claimed the top position on the Hugging Face OpenASR leaderboard with a word error rate of 6.14%, outperforming specialized speech recognition systems like WhisperV3. It also demonstrates competitive performance on vision tasks like mathematical and scientific reasoning with images.<\/p>\n\n\n\n Despite its compact size, Phi-4-mini demonstrates exceptional capabilities in text-based tasks. Microsoft reports the model \u201coutperforms similar size models and is on-par with models twice [as large]\u201d across various language-understanding benchmarks.<\/p>\n\n\n\n Particularly notable is the model\u2019s performance on math and coding tasks. According to the research paper, \u201cPhi-4-Mini consists of 32 Transformer layers with hidden state size of 3,072\u201d and incorporates group query attention to optimize memory usage for long-context generation.<\/p>\n\n\n\n On the GSM-8K math benchmark, Phi-4-mini achieved an 88.6% score, outperforming most 8-billion-parameter models, while on the MATH benchmark it reached 64%, substantially higher than similar-sized competitors.<\/p>\n\n\n\n \u201cFor the Math benchmark, the model outperforms similar sized models with large margins, sometimes more than 20 points. It even outperforms two times larger models\u2019 scores,\u201d the technical report notes.<\/p>\n\n\n\n Capacity, an AI \u201canswer engine\u201d that helps organizations unify diverse datasets, has already leveraged the Phi family to enhance its platform\u2019s efficiency and accuracy.<\/p>\n\n\n\n Steve Frederickson, head of product at Capacity, said in a statement, \u201cFrom our initial experiments, what truly impressed us about the Phi was its remarkable accuracy and the ease of deployment, even before customization. Since then, we\u2019ve been able to enhance both accuracy and reliability, all while maintaining the cost-effectiveness and scalability we valued from the start.\u201d<\/p>\n\n\n\n Capacity reported a 4.2x cost savings compared to competing workflows while achieving the same or better qualitative results for preprocessing tasks.<\/p>\n\n\n\n For years, AI development has been driven by a singular philosophy: bigger is better \u2014 more parameters, larger models, greater computational demands. But Microsoft\u2019s Phi-4 models challenge that assumption, proving that power isn\u2019t just about scale \u2014 it\u2019s about efficiency.<\/p>\n\n\n\n Phi-4-multimodal and Phi-4-mini are designed not for the data centers of tech giants, but for the real world \u2014 where computing power is limited, privacy concerns are paramount, and AI needs to work seamlessly without a constant connection to the cloud. These models are small, but they carry weight. Phi-4-multimodal integrates speech, vision and text processing into a single system without sacrificing accuracy, while Phi-4-mini delivers math, coding and reasoning performance on par with models twice its size.<\/p>\n\n\n\n This isn\u2019t just about making AI more efficient; it\u2019s about making it more accessible. Microsoft has positioned Phi-4 for widespread adoption, making it available through Azure AI Foundry, Hugging Face and the Nvidia API Catalog. The goal is clear: AI that isn\u2019t locked behind expensive hardware or massive infrastructure, but rather can operate on standard devices, at the edge of networks and in industries where compute power is scarce.<\/p>\n\n\n\n Masaya Nishimaki, a director at the Japanese AI firm Headwaters Co., Ltd., sees the impact firsthand. \u201cEdge AI demonstrates outstanding performance even in environments with unstable network connections or where confidentiality is paramount,\u201d he said in a statement. That means AI that can function in factories, hospitals, autonomous vehicles \u2014 places where real-time intelligence is required, but where traditional cloud-based models fall short.<\/p>\n\n\n\n At its core, Phi-4 represents a shift in thinking. AI isn\u2019t just a tool for those with the biggest servers and the deepest pockets. It\u2019s a capability that, if designed well, can work anywhere, for anyone. The most revolutionary thing about Phi-4 isn\u2019t what it can do \u2014 it\u2019s where it can do it.<\/p>\n Daily insights on business use cases with VB Daily<\/strong><\/p>\n If you want to impress your boss, VB Daily has you covered. We give you the inside scoop on what companies are doing with generative AI, from regulatory shifts to practical deployments, so you can share insights for maximum ROI.<\/p>\n \n\t\t\t\t\tThanks for subscribing. Check out more VB newsletters here.\n\t\t\t\t<\/p>\n An error occured.<\/p>\n<\/p><\/div>\n
\n<\/div>How Microsoft built a small AI model that does it all<\/h2>\n\n\n\n
Compact AI, massive impact: Phi-4-mini sets new performance standards<\/h2>\n\n\n\n
Transformative deployments: Phi-4\u2019s real-world efficiency in action<\/h2>\n\n\n\n
AI without limits: Microsoft\u2019s Phi-4 models bring advanced intelligence anywhere<\/h2>\n\n\n\n
![\"\"\/](\"https:\/\/venturebeat.com\/wp-content\/themes\/vb-news\/brand\/img\/vb-daily-phone.png\")
\n\t\t\t\t<\/div>\n<\/p><\/div>\n<\/div>\t\t\t<\/div>\r\n