This repository contains the full-merged 16-bit weights fine-tuned from Qwen/Qwen3-4B-Instruct-2507 using LoRA + Unsloth. No adapter loading is required.

Usage

Since this is a merged model, you can use it directly with transformers.

python

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "hara-CU/Advanced_FinalCandidate_482"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.float16,
    device_map="auto",
)

Sources & Terms (IMPORTANT)

Training data: hara-CU/LLM2025_DB_base_AW_345NoEAd_ALFformat_QH5L4R5_1392

Dataset License: MIT License. This dataset is used and distributed under the terms of the MIT License. Compliance: Users must comply with the MIT license (including copyright notice) and the base model's original terms of use.