---
base_model: unsloth/phi-4-unsloth-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- gguf
license: apache-2.0
language:
- en
datasets:
- bespokelabs/Bespoke-Stratos-17k
- bespokelabs/Bespoke-Stratos-35k
- NovaSky-AI/Sky-T1_data_17k
- Quazim0t0/BenfordsLawReasoningJSON
- open-thoughts/OpenThoughts-114k
---

# Uploaded  model

- **Developed by:** Quazim0t0
- **Finetuned from model :** unsloth/phi-4-unsloth-bnb-4bit
- **GGUF**
- **Trained for 8 Hours on A800 with the Bespoke Stratos 17k Dataset.**
- **Trained for 6 Hours on A800 with the Bespoke Stratos 35k Dataset.**
- **Trained for 2 Hours on A800 with the Benford's Law Reasoning Small 430 Row Dataset, ensuring no overfitting.**
- **Trained for 4 Hours on A800 with the Sky-T1_data_17k Dataset**
- **Trained for 2 Hours on A800 with the Openthoughts 114k Dataset.**
- **15$ Training...I'm actually amazed by the results.**


If using this model for Open WebUI here is a simple function to organize the models responses: https://openwebui.com/f/quaz93/phi4_turn_r1_distill_thought_function_v1

# Phi4 Turn R1Distill LoRA Adapters

## Overview
These **LoRA adapters** were trained using diverse **reasoning datasets** that incorporate structured **Thought** and **Solution** responses to enhance logical inference. This project was designed to **test the R1 dataset** on **Phi-4**, aiming to create a **lightweight, fast, and efficient reasoning model**.

All adapters were fine-tuned using an **NVIDIA A800 GPU**, ensuring high performance and compatibility for continued training, merging, or direct deployment.  
As part of an open-source initiative, all resources are made **publicly available** for unrestricted research and development.

---

## LoRA Adapters  
Below are the currently available LoRA fine-tuned adapters (**as of January 30, 2025**):

- [Phi4.Turn.R1Distill-Lora1](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill-Lora1)  
- [Phi4.Turn.R1Distill-Lora2](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill-Lora2)  
- [Phi4.Turn.R1Distill-Lora3](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill-Lora3)  
- [Phi4.Turn.R1Distill-Lora4](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill-Lora4)  
- [Phi4.Turn.R1Distill-Lora5](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill-Lora5)  
- [Phi4.Turn.R1Distill-Lora6](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill-Lora6)  
- [Phi4.Turn.R1Distill-Lora7](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill-Lora7)  
- [Phi4.Turn.R1Distill-Lora8](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill-Lora8)  

---

## GGUF Full & Quantized Models  
To facilitate broader testing and real-world inference, **GGUF Full and Quantized versions** have been provided for evaluation on **Open WebUI** and other LLM interfaces.

### **Version 1**
- [Phi4.Turn.R1Distill.Q8_0](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill.Q8_0)  
- [Phi4.Turn.R1Distill.Q4_k](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill.Q4_k)  
- [Phi4.Turn.R1Distill.16bit](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill.16bit)  

### **Version 1.1**
- [Phi4.Turn.R1Distill_v1.1_Q4_k](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill_v1.1_Q4_k)  

### **Version 1.2**
- [Phi4.Turn.R1Distill_v1.2_Q4_k](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill_v1.2_Q4_k)  

### **Version 1.3**
- [Phi4.Turn.R1Distill_v1.3_Q4_k-GGUF](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill_v1.3_Q4_k-GGUF)  

### **Version 1.4**
- [Phi4.Turn.R1Distill_v1.4_Q4_k-GGUF](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill_v1.4_Q4_k-GGUF)  

### **Version 1.5**
- [Phi4.Turn.R1Distill_v1.5_Q4_k-GGUF](https://huggingface.co./Quazim0t0/Phi4.Turn.R1Distill_v1.5_Q4_k-GGUF)  

---

## Usage  

### **Loading LoRA Adapters with `transformers` and `peft`**
To load and apply the LoRA adapters on Phi-4, use the following approach:

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base_model = "microsoft/Phi-4"
lora_adapter = "Quazim0t0/Phi4.Turn.R1Distill-Lora1"

tokenizer = AutoTokenizer.from_pretrained(base_model)
model = AutoModelForCausalLM.from_pretrained(base_model)
model = PeftModel.from_pretrained(model, lora_adapter)

model.eval()