Introducing Phi-4: Microsoft’s Newest Small Language Model …
Dec 13, 2024 · Learn about Phi-4, the latest small language model in Phi family, that offers high quality results at a small size (14B parameters).
Microsoft Phi-4 Compact AI Model Now Fully Open-Source
3 days ago · Source: Microsoft. In recent benchmarks, Phi-4 scored an impressive 91.8 on the American Mathematics Competition (AMC 12), outperforming Gemini Pro 1.5’s score of 89.8 and GPT-4o’s 77.9.
Microsoft launches Phi-4, a new generative AI model, in research ...
Dec 12, 2024 · Microsoft has revealed the newest addition to its Phi family of generative AI models. Called Phi-4, the model improves in several areas over its predecessors, Microsoft claims, particularly in...
Microsoft’s Phi-4 AI Model Goes Open Source! - It's FOSS News
3 days ago · Phi-4 Goes Open: What To Expect? Phi-4 on Hugging Face. Distributed under the MIT License, Phi-4 is a state-of-the-art 14 billion parameter SML that has been tailored for carrying complex reasoning tasks, outperforming popular models like Llama 3.3 and GPT-4o mini in benchmarks like MATH and GPQA.. Initially exclusive to Azure AI Foundry, Phi-4 is now available to everyone, offering a powerful ...
Microsoft AI Introduces Phi-4: A New 14 Billion Parameter Small ...
Dec 12, 2024 · To address these challenges, Microsoft Research has developed Phi-4, a 14-billion parameter language model that excels in reasoning tasks while being resource-efficient. Building on the Phi model family, Phi-4 incorporates novel approaches in synthetic data generation, curriculum design, and post-training refinement.
Microsoft introduces Phi-4, a 14B parameter state-of-the-art small ...
Dec 13, 2024 · Microsoft's new Phi-4 small language model outperforms OpenAI's GPT-4 and Google's Gemini 1.5 Pro in math-related AI benchmarks.
Microsoft makes Phi-4 model fully open source on Hugging Face
3 days ago · Phi-4 challenges the prevailing trend of scaling AI models to massive sizes. It demonstrates that smaller, well-designed models can achieve comparable or superior results in key areas.
We present phi-4, a 14-billion parameter language model developed with a training recipe that is centrally focused on data quality. Unlike most language models, where pre-training is based pri-marily on organic data sources such as web content or code, phi-4 strategically incorporates synthetic data throughout the training process.
Microsoft makes its Phi-4 small language model open-source
3 days ago · Microsoft said Phi-4 performed better on the popular GPQA and MATH benchmarks. These two test datasets contain science questions and math problems, respectively. More and more language models open-source. Phi-4 joins the growing list of small language models open-sourced by major technology companies over the past year.
[2412.08905] Phi-4 Technical Report - arXiv.org
Dec 12, 2024 · We present phi-4, a 14-billion parameter language model developed with a training recipe that is centrally focused on data quality. Unlike most language models, where pre-training is based primarily on organic data sources such as web content or code, phi-4 strategically incorporates synthetic data throughout the training process.
- Some results have been removed