Spaces:
Sleeping
Sleeping
File size: 2,230 Bytes
9bb3cc5 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 |
# =============================
# app.py for Cass Beta 2 Chat
# =============================
import gradio as gr
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
# =============================
# MODEL SETTINGS
# =============================
BASE_MODEL_ID = "ibm-granite/granite-4.0-micro-base"
PEFT_MODEL_ID = "DSDUDEd/Cass-Beta2.0"
print("🚀 Loading base model and PEFT adapter...")
tokenizer = AutoTokenizer.from_pretrained(BASE_MODEL_ID)
# Load base model
base_model = AutoModelForCausalLM.from_pretrained(BASE_MODEL_ID)
# Load PEFT model
model = PeftModel.from_pretrained(base_model, PEFT_MODEL_ID)
# Move to device
device = "cuda" if torch.cuda.is_available() else "cpu"
model.to(device)
model.eval()
print(f"✅ Model loaded on {device.upper()}")
# =============================
# CHAT HISTORY
# =============================
history = []
# =============================
# GENERATION FUNCTION
# =============================
def chat(user_input):
"""Generates AI response given user input"""
global history
history.append(("User", user_input))
# Prepare input
inputs = tokenizer(user_input, return_tensors="pt").to(device)
# Generate output
outputs = model.generate(
**inputs,
max_new_tokens=150,
pad_token_id=tokenizer.eos_token_id
)
ai_response = tokenizer.decode(outputs[0], skip_special_tokens=True)
history.append(("Cass", ai_response))
# Format chat
chat_text = ""
for role, message in history:
chat_text += f"{role}: {message}\n\n"
return chat_text
# =============================
# GRADIO INTERFACE
# =============================
with gr.Blocks() as demo:
gr.Markdown("## Chat with Cass Beta 2 🤖")
chatbox = gr.Textbox(label="Your message", placeholder="Type your message here...", lines=2)
send_button = gr.Button("Send")
output = gr.Textbox(label="Chat History", interactive=False, lines=20)
send_button.click(chat, inputs=chatbox, outputs=output)
chatbox.submit(chat, inputs=chatbox, outputs=output)
# =============================
# RUN APP
# =============================
if __name__ == "__main__":
demo.launch()
|