Update app.py

app.py

@@ -1,51 +1,47 @@
 import gradio as gr
 import os
-from transformers import pipeline, AutoTokenizer
+from huggingface_hub import InferenceClient
 
-# Load the tokenizer and model using the pipeline
-pipe = pipeline("text-generation", model="explorewithai/Loxa-4B", trust_remote_code=True)
-tokenizer = AutoTokenizer.from_pretrained("explorewithai/Loxa-4B")
+client = InferenceClient("explorewithai/Loxa-1.6B")
 
-# Get the system prompt from environment variables
 meo_system = os.environ.get("MEO")
 
 def respond(
     message,
-    history,
+    history: list[tuple[str, str]],
     max_tokens,
     temperature,
     top_p,
 ):
-    # Format the messages for the pipeline
     messages = [{"role": "system", "content": meo_system}]
-    for user_msg, bot_msg in history:
-        messages.append({"role": "user", "content": user_msg})
-        messages.append({"role": "assistant", "content": bot_msg})
+
+    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})
 
-    # Generate the prompt using the tokenizer's chat template
-    prompt = tokenizer.apply_chat_template(messages, tokenize=False)
+    response = ""
 
-    # Generate the response using the pipeline
-    outputs = pipe(
-        prompt,
-        max_new_tokens=max_tokens,
-        do_sample=True,
+    for message in client.chat_completion(
+        messages,
+        max_tokens=max_tokens,
+        stream=True,
         temperature=temperature,
         top_p=top_p,
-        return_full_text=False  # We only want the generated part
-    )
+    ):
+        token = message.choices[0].delta.content
 
-    # Extract the generated text
-    response = outputs[0]['generated_text']
+        response += token
+        yield response
 
-    return response
 
-# Create the Gradio interface
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
+        gr.Slider(minimum=1, maximum=2048, value=2048, step=1, label="Max new tokens"),
         gr.Slider(minimum=0.1, maximum=1.0, value=0.7, step=0.1, label="Temperature"),
         gr.Slider(
             minimum=0.1,
@@ -57,5 +53,6 @@ demo = gr.ChatInterface(
     ],
 )
 
+
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()