Update app.py

app.py

@@ -1,71 +1,64 @@
-import os
-import torch
-from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
 import gradio as gr
+from huggingface_hub import InferenceClient
 
-# Retrieve Hugging Face API key from environment variable
-hf_api_key = os.getenv('HF_API_KEY')
-if not hf_api_key:
-    raise ValueError("Hugging Face API key not found. Please set the 'HF_API_KEY' environment variable.")
+"""
+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("HuggingFaceH4/zephyr-7b-beta")
 
-# Configure bitsandbytes for 4-bit quantization
-bnb_config = BitsAndBytesConfig(
-    load_in_4bit=True,
-    bnb_4bit_use_double_quant=True,
-    bnb_4bit_quant_type="nf4",
-    bnb_4bit_compute_dtype=torch.bfloat16
-)
 
-# Load the tokenizer and model
-model_name = 'meta-llama/Llama-3.1-8B-Instruct'
-tokenizer = AutoTokenizer.from_pretrained(model_name, token=hf_api_key)
-model = AutoModelForCausalLM.from_pretrained(
-    model_name,
-    device_map='auto',
-    quantization_config=bnb_config,
-    token=hf_api_key
-)
+def respond(
+    message,
+    history: list[tuple[str, str]],
+    system_message,
+    max_tokens,
+    temperature,
+    top_p,
+):
+    messages = [{"role": "system", "content": system_message}]
 
-# Function to generate responses
-def generate_response(message, history):
-    try:
-        # Prepare the conversation history
-        conversation = ""
-        for user_msg, assistant_msg in history:
-            conversation += f"User: {user_msg}\nAssistant: {assistant_msg}\n"
-        conversation += f"User: {message}\nAssistant:"
+    for val in history:
+        if val[0]:
+            messages.append({"role": "user", "content": val[0]})
+        if val[1]:
+            messages.append({"role": "assistant", "content": val[1]})
 
-        # Tokenize the conversation
-        inputs = tokenizer(conversation, return_tensors="pt").to(model.device)
+    messages.append({"role": "user", "content": message})
 
-        # Generate response
-        outputs = model.generate(
-            **inputs,
-            max_new_tokens=150,
-            do_sample=True,
-            temperature=0.7,
-            top_p=0.9,
-            repetition_penalty=1.2
-        )
+    response = ""
 
-        # Decode the generated text
-        response = tokenizer.decode(outputs[0], skip_special_tokens=True).strip()
+    for message in client.chat_completion(
+        messages,
+        max_tokens=max_tokens,
+        stream=True,
+        temperature=temperature,
+        top_p=top_p,
+    ):
+        token = message.choices[0].delta.content
 
-        # Extract the assistant's response
-        response = response.split("Assistant:")[-1].strip()
+        response += token
+        yield response
 
-        return response
-    except Exception as e:
-        print("Error during response generation:", str(e))
-        return "An error occurred during response generation."
 
-# Create the Gradio interface
-chat_interface = gr.ChatInterface(
-    fn=generate_response,
-    title="Llama 3 Chatbot",
-    description="Interact with the Llama 3 model. Type a message and receive a response.",
-    examples=["Hello, how are you?", "What's the weather like?"]
+"""
+For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
+"""
+demo = gr.ChatInterface(
+    respond,
+    additional_inputs=[
+        gr.Textbox(value="You are a friendly Chatbot.", 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)",
+        ),
+    ],
 )
 
-# Launch the interface
-chat_interface.launch(debug=True)
+
+if __name__ == "__main__":
+    demo.launch()