fixed maybe??

app.py

@@ -1,12 +1,12 @@
 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("Electricarchmage/cookbookgpt")
+from transformers import GPT2LMHeadModel, GPT2Tokenizer
 
+# Load model and tokenizer from Hugging Face Hub
+model_name = "Electricarchmage/cookbookgpt"
+model = GPT2LMHeadModel.from_pretrained(model_name)
+tokenizer = GPT2Tokenizer.from_pretrained(model_name)
 
+# Define the respond function
 def respond(
     message,
     history: list[tuple[str, str]],
@@ -15,34 +15,39 @@ def respond(
     temperature,
     top_p,
 ):
-    messages = [{"role": "system", "content": system_message}]
-
+    # Preparing the messages for context (the history and the new message)
+    input_text = system_message + "\n"
+    
     for val in history:
         if val[0]:
-            messages.append({"role": "user", "content": val[0]})
+            input_text += f"User: {val[0]}\n"
         if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+            input_text += f"Assistant: {val[1]}\n"
 
-    messages.append({"role": "user", "content": message})
+    input_text += f"User: {message}\nAssistant:"
 
-    response = ""
+    # Tokenize the input and generate a response
+    inputs = tokenizer(input_text, return_tensors="pt", max_length=1024, truncation=True)
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
+    # Generate output tokens
+    output = model.generate(
+        inputs["input_ids"],
+        max_length=max_tokens + len(inputs["input_ids"][0]),
         temperature=temperature,
         top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+        num_return_sequences=1,
+        no_repeat_ngram_size=2,
+    )
 
-        response += token
-        yield response
+    # Decode the output tokens into text
+    response = tokenizer.decode(output[0], skip_special_tokens=True)
 
+    # Extract only the assistant's reply
+    assistant_reply = response.split("Assistant:")[-1].strip()
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+    return assistant_reply
+
+# Define the Gradio interface
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
@@ -59,6 +64,6 @@ demo = gr.ChatInterface(
     ],
 )
 
-
+# Launch the app
 if __name__ == "__main__":
     demo.launch()