Update app.py

app.py

@@ -1,40 +1,44 @@
 import gradio as gr
+import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
 
-
-# Replace "username/llama3.3" with your actual model repository
 MODEL_NAME = "tiiuae/falcon-7b-instruct"
 
 # Load tokenizer and model
 tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
 model = AutoModelForCausalLM.from_pretrained(MODEL_NAME, torch_dtype="auto")
 
+# If your model doesn't define a pad token, you can use the eos token instead:
+if tokenizer.pad_token is None:
+    tokenizer.pad_token = tokenizer.eos_token
+if model.config.pad_token_id is None:
+    model.config.pad_token_id = tokenizer.eos_token_id
+
 # Create a text-generation pipeline
 text_gen = pipeline(
     "text-generation",
     model=model,
     tokenizer=tokenizer,
     max_length=512,
+    truncation=True,  # <-- Explicitly enable truncation
     do_sample=True,
     temperature=0.7
 )
 
 def chat(user_input):
-    """
-    Simple chat function that prepends user input to a system prompt (if needed)
-    and returns the model's text generation.
-    """
-    # If you have a special prompt format for a chat model, add it here.
-    # For a generic chat, you can just send the user_input:
-    outputs = text_gen(user_input, max_length=512)
+    outputs = text_gen(
+        user_input, 
+        max_length=512,
+        truncation=True  # <-- Also ensure truncation is True here
+    )
     return outputs[0]["generated_text"]
 
 demo = gr.Interface(
     fn=chat,
     inputs="text",
     outputs="text",
-    title="LLaMA3.3 Chat (Example)",
-    description="A chat interface for the LLaMA-based model named 'llama3.3'."
+    title="Falcon-7B-Instruct Chat (Example)",
+    description="A chat interface for Falcon-7B-Instruct."
 )
 
 if __name__ == "__main__":