Instructions to use mihirpatel1111/ReasonForge-3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mihirpatel1111/ReasonForge-3B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="mihirpatel1111/ReasonForge-3B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mihirpatel1111/ReasonForge-3B")
model = AutoModelForCausalLM.from_pretrained("mihirpatel1111/ReasonForge-3B", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mihirpatel1111/ReasonForge-3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mihirpatel1111/ReasonForge-3B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mihirpatel1111/ReasonForge-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mihirpatel1111/ReasonForge-3B

SGLang

How to use mihirpatel1111/ReasonForge-3B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "mihirpatel1111/ReasonForge-3B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mihirpatel1111/ReasonForge-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "mihirpatel1111/ReasonForge-3B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mihirpatel1111/ReasonForge-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use mihirpatel1111/ReasonForge-3B with Docker Model Runner:
```
docker model run hf.co/mihirpatel1111/ReasonForge-3B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

# ReasonForge-3B: Fine-Tuning Llama-3.2-3B-Instruct

This project details the complete end-to-end pipeline for fine-tuning, evaluating, exporting, and running a reasoning model using Unsloth, QLoRA, and Ollama.

🚀 Project Overview

Base Model: unsloth/Llama-3.2-3B-Instruct (Quantized 4-bit)
Fine-Tuned Model: ReasonForge-3B
Dataset: ServiceNow-AI/R1-Distill-SFT (v0)
Inference & Deployment Engine: Ollama (using GGUF conversion)

🛠️ Step-by-Step Workflow

1. Installation & Environment Setup

We initialize our environment by installing Unsloth, which accelerates training by up to 2x and significantly reduces memory usage:

pip install unsloth
pip install --force-reinstall --no-cache-dir --no-deps git+https://github.com/unslothai/unsloth.git

2. Loading the Model & Tokenizer

We load a 4-bit quantized version of the Llama-3.2-3B model using Unsloth's FastLanguageModel class:

from unsloth import FastLanguageModel

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "unsloth/Llama-3.2-3B-Instruct",
    max_seq_length = 2048,
    dtype = None,
    load_in_4bit = True,
)

3. Applying PEFT/LoRA Adapters

To optimize parameter-efficient fine-tuning (QLoRA), we attach LoRA adapters onto targeted linear projection layers:

model = FastLanguageModel.get_peft_model(
    model,
    r = 16,
    target_modules = ["q_proj", "k_proj", "v_proj", "o_proj", "gate_proj", "up_proj", "down_proj"],
    lora_alpha = 16,
    lora_dropout = 0,
    bias = "none",
    use_gradient_checkpointing = "unsloth",
    random_state = 3407,
)

4. Dataset Structuring & Fine-Tuning

We map our raw reasoning data using an custom r1_prompt template and train the adapter using SFTTrainer for 30 steps:

trainer_stats = trainer.train()

5. Native Inference & GGUF Conversion

We test the model natively after patching the chat template for inference. The fine-tuned weights are saved locally and exported directly to a GGUF format Llama-3.2-3B-Instruct.Q8_0.gguf using Unsloth's packaging features:

model.save_pretrained_gguf("ReasonForge-3BF", tokenizer)

6. Local Running via Ollama

We set up and launch Ollama within Colab, load the generated GGUF file via a custom Modelfile, and run the model natively through the command line:

# Creating and launching the ollama service
ollama create reasonforge -f ReasonForge-3BF_gguf/Modelfile
ollama run reasonforge "How many r's are in strawberry?"