Update app.py

app.py

@@ -91,36 +91,16 @@ class LocalInferenceClient:
 
 # Specify the model paths for gguf models
 model_configs = {
-    "Lake 1 Chat": {
-        "path": r"C:\Users\BI Corp\Videos\main\Lake-1-chat\Lake-1-Chat.gguf",
+    "Test": {
+        "path": r"./test-model.gguf",
         "specs": """
             ## Lake 1 Chat Specifications
-            - **Architecture**: Lake 1
-            - **Parameters**: 14B
-            - **Capabilities**: Reasoning, logical inference, coding
-            - **Intended Use**: Suitable for complex reasoning tasks, math, coding problems, and detailed conversations.
+            - **Architecture**: Test
+            - **Parameters**: IDK
+            - **Capabilities**: test
+            - **Intended Use**: test
         """
-    },
-    "Lake 1 Mini": {
-        "path": r"C:\Users\BI Corp\Videos\main\Lake-1-mini\Lake-1-Mini.gguf",
-        "specs": """
-            ## Lake 1 Mini Specifications
-            - **Architecture**: Lake 1
-            - **Parameters**: 6B
-            - **Capabilities**: Quick responses, compact model
-            - **Intended Use**: Great for fast responses and lightweight use cases.
-        """
-    },
-    "Lake 1 Base": {
-        "path": r"C:\Users\BI Corp\Videos\main\Lake-1-base\Lake-1-Base.gguf",
-        "specs": """
-            ## Lake 1 Base Specifications
-            - **Architecture**: Lake 1
-            - **Parameters**: 12B
-            - **Capabilities**: Balanced performance between speed and accuracy
-            - **Intended Use**: Best for use cases requiring a balance of speed and detail in responses.
-        """
-    },
+    }
 }
 
 # Set up a dictionary mapping model names to their clients
@@ -128,17 +108,7 @@ clients = {name: LocalInferenceClient(name, config['path']) for name, config in
 
 # Presets for performance/quality tradeoffs
 presets = {
-    "Lake 1 Mini": {
-        "Fast": {"max_new_tokens": 100, "temperature": 1.0, "top_p": 0.9},
-        "Normal": {"max_new_tokens": 200, "temperature": 0.7, "top_p": 0.95},
-        "Quality": {"max_new_tokens": 300, "temperature": 0.5, "top_p": 0.90},
-    },
-    "Lake 1 Base": {
-        "Fast": {"max_new_tokens": 100, "temperature": 1.0, "top_p": 0.9},
-        "Normal": {"max_new_tokens": 200, "temperature": 0.7, "top_p": 0.95},
-        "Quality": {"max_new_tokens": 300, "temperature": 0.5, "top_p": 0.90},
-    },
-    "Lake 1 Chat": {
+    "Test": {
         "Fast": {"max_new_tokens": 100, "temperature": 1.0, "top_p": 0.9},
         "Normal": {"max_new_tokens": 200, "temperature": 0.7, "top_p": 0.95},
         "Quality": {"max_new_tokens": 300, "temperature": 0.5, "top_p": 0.90},
@@ -147,10 +117,8 @@ presets = {
 
 # A system prompt for the model
 system_messages = {
-    "Lake 1 Chat": "You are Lake 1 Chat, a powerful open-source reasoning model. Think carefully and answer step by step.",
-    "Lake 1 Mini": "You are Lake 1 Mini, a powerful open-source compact model. Think and answer fast.",
-    "Lake 1 Base": "You are Lake 1 Base, a powerful open-source original model. Think and answer step by step but balance speed and accuracy.",
-}
+    "Test": "You are Lake 1 Chat, a powerful open-source reasoning model. Think carefully and answer step by step.",
+    }
 
 def generate_response(message: str, model_name: str, preset: str) -> str:
     """
@@ -212,7 +180,7 @@ with gr.Blocks(title="BI CORP AI Assistant", theme="soft") as demo:
                 interactive=True
             )
             model_info_md = gr.Markdown(
-                value=model_configs["Lake 1 Chat"]["specs"],
+                value=model_configs["Test"]["specs"],
                 label="📝 Model Specifications"
             )
         
@@ -220,7 +188,7 @@ with gr.Blocks(title="BI CORP AI Assistant", theme="soft") as demo:
             chat_interface = gr.ChatInterface(
                 fn=handle_chat,
                 additional_inputs=[model_dropdown, preset_dropdown],
-                examples=[["Explain quantum computing", "Lake 1 Chat", "Normal"]],
+                examples=[["Explain quantum computing", "Test", "Normal"]],
                 chatbot=gr.Chatbot(height=600, label="💬 Conversation", show_copy_button=True),
                 textbox=gr.Textbox(placeholder="Type your message...", container=False, scale=7, autofocus=True),
                 submit_btn=gr.Button("🚀 Send", variant="primary")