# Mathematical Code Optimization
Deepseek R1 Zero
MIT
DeepSeek-R1 is the first-generation reasoning model developed by DeepSeek, trained through reinforcement learning, excelling in mathematics, coding, and reasoning tasks.
Large Language Model
Transformers

D
deepseek-ai
4,034
905
Phi 4
MIT
Phi-4 is an open-source language model developed by Microsoft Research, focusing on high-quality data and advanced reasoning capabilities, with 14 billion parameters.
Large Language Model
Transformers Supports Multiple Languages

P
microsoft
418.44k
2,011
Featured Recommended AI Models