final

app.py

@@ -2,7 +2,11 @@ import gradio as gr
 from huggingface_hub import InferenceClient
 import os
 
+# Securely retrieve the token from your Space secrets
+# Ensure you have a secret named HF_TOKEN in your Settings
 HF_TOKEN = os.getenv("HF_TOKEN")
+
+# Initialize the inference client with the specified model
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta", token=HF_TOKEN)
 
 class StateController:
@@ -33,8 +37,8 @@ class StateController:
         self.state_array = [0] * 121
         return "System resolved. State array reset to zero."
 
-def process_request(message, history):
-    # Hardware override sequence
+def generate_response(message, history):
+    # Hardware diagnostic override
     if "run grid diagnostic" in message.lower():
         controller = StateController()
         output = "Diagnostic sequence initiated.\n\n"
@@ -43,54 +47,44 @@ def process_request(message, history):
         output += f"`{controller.render_grid()}`\n\n"
         output += "Executing state resolution:\n"
         output += f"`{controller.resolve_grid()}`"
-        yield output
-        return
+        return output
 
     system_instruction = (
         "You are a logic-focused inference engine. "
-        "You utilize strict state-hold memory and parallel integer blocks to process queries. "
-        "Provide highly technical, accurate, and direct responses."
+        "You utilize strict state-hold memory and parallel integer blocks. "
+        "Provide direct, technical, and accurate responses."
     )
     
-    messages = [{"role": "system", "content": system_instruction}]
-    for human, assistant in history:
-        messages.append({"role": "user", "content": human})
-        messages.append({"role": "assistant", "content": assistant})
-    messages.append({"role": "user", "content": message})
+    # Formatting for Gradio 6.5+ message history
+    formatted_messages = [{"role": "system", "content": system_instruction}]
+    for turn in history:
+        formatted_messages.append({"role": "user", "content": turn[0]})
+        formatted_messages.append({"role": "assistant", "content": turn[1]})
+    formatted_messages.append({"role": "user", "content": message})
 
-    response_text = ""
     try:
-        for chunk in client.chat_completion(
-            messages,
+        response_text = ""
+        # Direct call for response generation
+        completion = client.chat_completion(
+            formatted_messages,
             max_tokens=1024,
-            stream=True,
-        ):
-            token = chunk.choices[0].delta.content
-            if token:
-                response_text += token
-                yield response_text
+            stream=False # Set to False for maximum stability during testing
+        )
+        return completion.choices[0].message.content
     except Exception as error:
-        error_message = f"Connection exception: {str(error)}. Verify API token permissions."
-        yield error_message
-
-custom_css = """
-body, .gradio-container { background-color: #0b0f19 !important; }
-footer {display: none !important}
-.message.user { background-color: #1e293b !important; border: 1px solid #3b82f6 !important; }
-.message.bot { background-color: #0f172a !important; color: #60a5fa !important; }
-"""
+        return f"System Error: {str(error)}. Ensure HF_TOKEN is correctly set in Secrets."
 
-with gr.Blocks(css=custom_css, theme=gr.themes.Soft(primary_hue="blue")) as demo:
+# Professional UI implementation
+with gr.Blocks(theme=gr.themes.Soft(primary_hue="blue")) as demo:
     gr.Markdown("# Advanced Logic Interface")
     gr.ChatInterface(
-        fn=process_request,
-        description="Inference layer utilizing strict grid logic.",
+        fn=generate_response,
+        description="Inference layer utilizing state-hold logic.",
         examples=[
             "Run grid diagnostic",
-            "Calculate allocation requirements for 120 units across 3 nodes.",
-            "Define processing latency without using the words delay or time."
-        ],
-        cache_examples=False
+            "Explain network latency without using the word delay.",
+            "Calculate allocation for 120 units across 3 nodes."
+        ]
     )
 
 if __name__ == "__main__":