app.py

# Step 1: Force install torch first (if not already installed)
import os
os.system("pip install torch")

# Step 2: Install autoawq after torch is available
os.system("pip install autoawq")

# Now import everything safely
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
import gradio as gr

model_id = "TheBloke/MelloGPT-AWQ"

# Load tokenizer and model
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    trust_remote_code=True,
    torch_dtype=torch.float16,
    device_map="auto"
)

def chat(user_input):
    inputs = tokenizer(user_input, return_tensors="pt").to(model.device)
    outputs = model.generate(**inputs, max_new_tokens=150, do_sample=True)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return response

gr.Interface(fn=chat, inputs="text", outputs="text", title="MelloGPT Mental Health Chatbot").launch()