app.py

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

MODEL_NAME = "NeuraCraft/Lance-AI-V2"

torch.backends.cudnn.deterministic = True
torch.backends.cudnn.benchmark = False

print("Loading model...")

tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
if tokenizer.pad_token is None:
    tokenizer.pad_token = tokenizer.eos_token

model = AutoModelForCausalLM.from_pretrained(
    MODEL_NAME,
    trust_remote_code=True
)
model.eval()

def generate(text):
    inputs = tokenizer(text, return_tensors="pt")

    with torch.no_grad():
        output = model.generate(
            **inputs,
            max_new_tokens=100,
            do_sample=True,
            temperature=0.7,
            top_p=0.9,
            repetition_penalty=1.2,
            pad_token_id=tokenizer.eos_token_id
        )

    return tokenizer.decode(output[0], skip_special_tokens=True)

demo = gr.Interface(
    fn=generate,
    inputs=gr.Textbox(label="Input"),
    outputs=gr.Textbox(label="Lance AI"),
    title="Lance AI V2"
)

demo.launch()