Update app.py

app.py

@@ -1,7 +1,7 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
 
-# Available open-source base models (completion style)
+# Available open-source base models
 MODEL_CHOICES = {
     "Mistral 7B Instruct (mistralai/Mistral-7B-Instruct-v0.2)": "mistralai/Mistral-7B-Instruct-v0.2",
     "Falcon 7B Instruct (tiiuae/falcon-7b-instruct)": "tiiuae/falcon-7b-instruct",
@@ -15,7 +15,7 @@ def complete_text(prompt, max_tokens, temperature, top_p, model_choice, hf_token
     Streams output token-by-token.
     """
     if not hf_token or not hf_token.token:
-        yield "⚠️ Please log in with your Hugging Face account (for gated models like LLaMA-2)."
+        yield "⚠️ Please log in with your Hugging Face account (needed for gated models like LLaMA-2)."
         return
 
     model_id = MODEL_CHOICES[model_choice]
@@ -31,9 +31,8 @@ def complete_text(prompt, max_tokens, temperature, top_p, model_choice, hf_token
         repetition_penalty=1.0,
     )
 
-    for event in stream:
-        # Each event is a string chunk
-        response_text += event
+    for chunk in stream:
+        response_text += chunk
         yield response_text
 
 
@@ -72,9 +71,10 @@ with gr.Blocks() as demo:
                 lines=15,
             )
 
+    # ✅ Pass gr.OAuthToken as an input type (no manual instantiation)
     submit.click(
         fn=complete_text,
-        inputs=[prompt, max_tokens, temperature, top_p, model_choice, gr.OAuthToken()],
+        inputs=[prompt, max_tokens, temperature, top_p, model_choice, gr.OAuthToken],
         outputs=output,
     )