non quantized update

app.py

@@ -1,34 +1,23 @@
 import gradio as gr
-from llama_cpp import Llama
-from huggingface_hub import hf_hub_download
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-# Fetch the model file from Hugging Face Hub
-model_repo = "SupermanRX/moderateTherapistModel"  # Replace with your repo name
-model_file = "unsloth.Q8_0.gguf"  # Replace with your GGUF model file name in the repo
+# Load your fine-tuned model
+model_name = "SupermanRX/therapistAi"  # Replace with your model's Hugging Face path
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name)
 
-# Download the model file to the local environment
-model_path = hf_hub_download(repo_id=model_repo, filename=model_file)
-
-# Load the GGUF model
-llm = Llama(model_path=model_path)
-
-# Define the chatbot function
 def chatbot(input_text):
-    output = llm(prompt=input_text, max_tokens=200)
-    return output['choices'][0]['text']
+    # Generate a response
+    inputs = tokenizer(input_text, return_tensors="pt")
+    outputs = model.generate(inputs["input_ids"], max_length=200, pad_token_id=tokenizer.eos_token_id)
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return response
 
-# Create Gradio interface
+# Use the default chatbot interface
 with gr.Blocks() as demo:
-    chatbot_ui = gr.Chatbot()
-    textbox = gr.Textbox(label="Type your message here:")
-    submit = gr.Button("Send")
-
-    # Handle user interaction
-    def user_interaction(input_text, chat_history):
-        response = chatbot(input_text)
-        chat_history.append((input_text, response))
-        return chat_history, ""
-
-    submit.click(user_interaction, [textbox, chatbot_ui], [chatbot_ui, textbox])
-
+    gr.Chatbot().style(height=600).chat(
+        chatbot,
+        placeholder="Type your message here...",
+        show_label=False
+    )
 demo.launch()