chhatramani/qwen3-4B-civillaw-en-v1-gguf

Nepal Civil Law Q&A (Qwen3-4B-Instruct Fine-Tuned)

NyayaLM is a specialized large language model fine-tuned for understanding and answering queries related to the Nepal Civil Code. It is based on the Qwen3-4B-Instruct architecture and has been optimized using Unsloth for efficient 4-bit LoRA (Low-Rank Adaptation) training.

The model is designed to assist in legal reasoning, providing information on civil rights, property laws, family laws, and legal procedures within the context of the Nepalese legal system.

Model Details

• Developed by: [Chhatramani Yadav]
• Base Model: unsloth/Qwen3-4B-Instruct-2507
• Language(s): English (Legal Domain)
• License: Apache-2.0
• Fine-tuning Technique: LoRA (Rank-Stabilized - RS-LoRA)
• Quantization: 4-bit (via Unsloth/bitsandbytes)

Intended Use

NyayaLM is intended for legal researchers, practitioners, and students in Nepal. It can:

• Answer specific questions based on the Nepal Civil Code.
• Provide legal context for various civil scenarios.
• Assist in drafting and summarizing legal principles.

Note: This model is for experimental purposes only and should not be treated as professional legal advice.

Training Data

The model was trained on a high-quality mixed dataset comprising two distinct types of legal QA:

1. Real-World Civil Code Legal QA (Type 2): 2,991 examples focusing on practical legal applications.
2. Conversational Civil Code QA (Type 3): 2,386 examples curated for multi-turn dialogue.

Dataset Statistics:

• Total Raw Samples: 5,377
• Mixing Ratio: 70% Real-World QA / 30% Conversational QA
• Final Mixed Dataset Size: 4,272 examples
• Split: 90% Training (3,844 samples), 10% Evaluation (428 samples)

Training Hyperparameters

The following configuration was used for fine-tuning via Unsloth:

Parameter	Value
LoRA Rank (r)	16
LoRA Alpha	32
LoRA Dropout	0.08
RS-LoRA	Enabled
Optimizer	AdamW (8-bit)
Learning Rate	1.2e-4
Max Sequence Length	2048
Weight Decay	0.01
Batch Size	2 (with Gradient Accumulation)
Precision	4-bit NormalFloat (NF4)

Evaluation Results

The model was evaluated on 20 specific test cases comparing original context vs. zero-context performance.

Fine-tune Model Evaluation Metrics

Metric	Score
ROUGE-1	0.3749
ROUGE-2	0.1459
ROUGE-L	0.2436
BERTScore (F1 Avg)	0.8808
BERTScore (With Context)	0.8808

Insights: The high BERTScore indicates strong semantic alignment with legal ground truths, even when lexical overlap (ROUGE) varies.

How to Use

You can run this model using the Unsloth library for 2x faster inference:

from unsloth import FastLanguageModel
import torch

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "your-username/NyayaLM-Nepal-Civil-Law",
    max_seq_length = 2048,
    load_in_4bit = True,
)
FastLanguageModel.for_inference(model)

# Example Prompt
messages = [
    {"role": "system", "content": "You are a legal assistant specializing in Nepal Civil Law."},
    {"role": "user", "content": "What is the time limit for registering a birth in Nepal?"}
]

inputs = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to("cuda")
outputs = model.generate(input_ids=inputs, max_new_tokens=512)
print(tokenizer.batch_decode(outputs))

qwen3-4B-civillaw-en-v1-gguf : GGUF

This model was finetuned and converted to GGUF format using Unsloth.

Example usage:

• For text only LLMs: ./llama.cpp/llama-cli -hf chhatramani/qwen3-4B-civillaw-en-v1-gguf --jinja
• For multimodal models: ./llama.cpp/llama-mtmd-cli -hf chhatramani/qwen3-4B-civillaw-en-v1-gguf --jinja

Available Model files:

• qwen3-4b-instruct-2507.Q4_K_M.gguf

Ollama

An Ollama Modelfile is included for easy deployment. This was trained 2x faster with Unsloth