Phi-4 Mini Reasoning

Microsoft

Chain-of-thought reasoning variant of Phi-4 Mini (3.8B). SFT on OpenAI o3-mini demonstrations + RL. 128K context makes it ideal for RAG. Best reasoning at sub-4B scale.

Text Generation Local Latest Phi Family v4
Documentation Back to Models
Parameters
3.8B
params
Context Window
131K
tokens
Max Output
-
tokens
Input Price
-
per 1M tokens
Output Price
-
per 1M tokens

Phi Family Timeline 6 versions

Phi-4 Mini Reasoning Current Latest local
Apr 2025
Phi-4 14B local
Dec 2024
Phi-3.5 Mini local
Aug 2024
Phi-3 Mini 3.8B local
Apr 2024
Phi-3 Small 7B local
Apr 2024
Phi-3 Medium 14B local
Apr 2024

Capabilities

👁️
Vision
⚡
Function Calling
📋
JSON Mode
🌊
Streaming
💬
System Prompt
🖥️
Code Execution
🔍
Web Search
🔌
MCP Support

Local Model Specs

Quantization
Q4_K_M
Architecture
Dense Transformer
Runtime
Ollama / llama.cpp
Disk Size
2.5 GB

Details

Release Date
April 30, 2025
Knowledge Cutoff
-
Source
Local
License
MIT
Model ID
phi-4-mini-reasoning
Last updated: March 13, 2026