DeepSeek-R1-Distill-phi-3-mini-4k Open-Source Inference Model

Home

Deepseek R1 Distill Phi 3 Mini 4k Lorar8 Alpha16 50000samples

Developed by GPD1

A reasoning model based on Deepseek-R1 knowledge distillation, supporting Chain-of-Thought (CoT) reasoning capabilities

Large Language Model

Safetensors

EnglishOpen Source License:MIT #Knowledge Distillation Inference #English CoT Generation #Phi-3-mini Optimization

Downloads 71

Release Time : 1/31/2025

Model Overview

This model is a reasoning model extracted through knowledge distillation from Deepseek-R1 and Llama-70B models, focusing on improving performance in complex reasoning tasks.

Model Features

Knowledge Distillation

Extracts knowledge from Deepseek-R1 and Llama-70B large models, reducing model size while maintaining high performance

Chain-of-Thought Reasoning

Supports CoT (Chain-of-Thought) reasoning capabilities, suitable for solving complex reasoning problems

Efficient Inference

Optimized based on Phi-3-mini architecture, improving inference efficiency while maintaining performance

Model Capabilities

Text generation

Complex logical reasoning

Knowledge Q&A

Chain-of-thought reasoning

Use Cases

Education

Mathematical Problem Solving

Solving mathematical problems requiring multi-step reasoning

Research

Scientific Reasoning

Assisting in reasoning and verification of scientific hypotheses

Property	Details
Model Type	Distilled model from Deepseek-R1 Knowledge
Base Model	microsoft/Phi-3-mini-4k-instruct
Training Data	Magpie-Align/Magpie-Reasoning-V2-250K-CoT-Deepseek-R1-Llama-70B
Pipeline Tag	text-generation
Tags	Deepseek, Distillation

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご

Deepseek R1 Distill Phi 3 Mini 4k Lorar8 Alpha16 50000samples

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 Distilled Model from Deepseek-R1 Knowledge

🚀 Quick Start

📄 License

📦 Model Information