Microsoft's Phi 4 Reasoning Plus, released in May 2025, is a 14B parameter model designed for advanced reasoning in math, science, and coding. This model is fine-tuned with reinforcement learning to enhance accuracy and provides detailed, step-by-step reasoning traces alongside final answers. With a context window of 32,768 tokens, it supports text-to-text interactions, making it ideal for English-only applications where detailed reasoning is crucial. While responses are more comprehensive, they may also take longer to generate, prioritizing quality over speed.
Use Cases
Here are a few ways teams apply Microsoft: Phi 4 Reasoning Plus in practice—from fast drafting to multimodal understanding. Adapt these ideas to your workflow.
Improve accuracy in math problem-solving
Enhance scientific reasoning tasks
Generate detailed coding explanations
Support structured reasoning workflows
Develop educational tools for complex topics
Key Features
A quick look at the capabilities that make this model useful in real projects.
14B parameter model for advanced reasoning
Dense decoder-only transformer architecture
Fine-tuned with reinforcement learning
32,768 token context window
Structured step-by-step reasoning outputs
English-only application focus
Specs
Overview
Vendor
microsoft
Model ID
microsoft/phi-4-reasoning-plus
Release
2025-05-01
Modalities & context
Input
text
Output
text
Context
32,768 tokens
Parameters & defaults
Supported parameters: frequency_penalty, include_reasoning, max_tokens, min_p, presence_penalty, reasoning, repetition_penalty, response_format, seed, stop, temperature, top_k, top_p
Defaults: temperature 0.2, top_p 0.95
Benchmark tests: Microsoft: Phi 4 Reasoning Plus
We ran this model against a few representative prompts to show its range. Review the outputs below and be the judge.
Text
Prompt:
Write 150 words on how AI might positively upend work, leisure and creativity
Artificial Intelligence (AI) has the potential to revolutionize work, leisure, and creativity in transformative ways. In the workplace, AI can automate routine and repetitive tasks, allowing employees to focus on more strategic and creative endeavors. This shift can lead to increased productivity and job satisfaction, as workers engage in more meaningful and fulfilling activities. AI-driven tools can also enhance decision-making processes by providing data-driven insights, enabling businesses to innovate and adapt more swiftly to market changes.
In terms of leisure, AI can personalize entertainment experiences, offering tailored recommendations for movies, music, and games based on individual preferences. This personalization can enhance user satisfaction and engagement, making leisure time more enjoyable and efficient. Moreover, AI can facilitate virtual and augmented reality experiences, providing immersive and interactive forms of entertainment that were previously unimaginable.
Creativity stands to benefit significantly from AI, as it can serve as a collaborative partner in the creative process. AI algorithms can generate novel ideas, assist in design and art creation, and even compose music or write poetry. This collaboration can push the boundaries of human creativity, leading to new forms of artistic expression and innovation. By augmenting human capabilities, AI can help unlock untapped creative potential, fostering a renaissance of creativity across various fields.
Run this prompt on Upend.AI