app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_id = "devNaam/vakilai-llama32-3b-v1"

tokenizer = AutoTokenizer.from_pretrained(model_id)

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

def generate_response(prompt):

    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)

    outputs = model.generate(
        **inputs,
        max_new_tokens=300,
        temperature=0.7,
        top_p=0.9
    )

    response = tokenizer.decode(outputs[0], skip_special_tokens=True)

    return response


iface = gr.Interface(
    fn=generate_response,
    inputs=gr.Textbox(lines=5, placeholder="Ask VakilAI a legal question..."),
    outputs="text",
    title="VakilAI Legal Assistant",
    description="AI Legal assistant trained on Indian legal data."
)

iface.launch()