Update app.py

app.py

@@ -1,11 +1,14 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient(model = "Braszczynski/Llama-3.2-3B-Instruct-bnb-4bit-460steps")
-
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Load the model and tokenizer
+model_name = "Braszczynski/Llama-3.2-3B-Instruct-bnb-4bit-460steps"
+tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=True)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    load_in_4bit=True,  # Ensure this matches your model's quantization
+    device_map="auto"   # Automatically allocate model layers to GPUs
+)
 
 def respond(
     message,
@@ -15,50 +18,39 @@ def respond(
     temperature,
     top_p,
 ):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
+    # Combine system message and chat history
+    chat_history = f"{system_message}\n"
+    for user_msg, bot_reply in history:
+        chat_history += f"User: {user_msg}\nAssistant: {bot_reply}\n"
+    chat_history += f"User: {message}\nAssistant:"
+
+    # Tokenize the input
+    inputs = tokenizer(chat_history, return_tensors="pt", truncation=True).to("cuda")
+
+    # Generate response
+    outputs = model.generate(
+        inputs["input_ids"],
+        max_new_tokens=max_tokens,
         temperature=temperature,
         top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+        pad_token_id=tokenizer.eos_token_id
+    )
 
-        response += token
-        yield response
+    # Decode and format the output
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    response = response[len(chat_history):].strip()  # Remove input context from output
+    return response
 
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# Define the Gradio interface
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
+        gr.Textbox(value="You are a friendly assistant.", label="System message"),
         gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
         gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
+        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
     ],
 )
 
-
 if __name__ == "__main__":
     demo.launch()