Update app.py

app.py

@@ -14,19 +14,7 @@ from pydantic import BaseModel
 load_dotenv()
 HUGGINGFACE_TOKEN = os.getenv("HUGGINGFACE_TOKEN")
 
-global_data = {
-    'models': {},
-    'tokens': {
-        'eos': 'eos_token',
-        'pad': 'pad_token',
-        'padding': 'padding_token',
-        'unk': 'unk_token',
-        'bos': 'bos_token',
-        'sep': 'sep_token',
-        'cls': 'cls_token',
-        'mask': 'mask_token'
-    }
-}
+global_data = {'models': {}, 'tokens': {'eos': 'eos_token', 'pad': 'pad_token', 'padding': 'padding_token', 'unk': 'unk_token', 'bos': 'bos_token', 'sep': 'sep_token', 'cls': 'cls_token', 'mask': 'mask_token'}}
 
 model_configs = [
     {"repo_id": "Hjgugugjhuhjggg/mergekit-ties-tzamfyy-Q2_K-GGUF", "filename": "mergekit-ties-tzamfyy-q2_k.gguf", "name": "my_model"}
@@ -44,11 +32,13 @@ def load_model(model_config):
             return model
         except Exception as e:
             print(f"Error loading model {model_name}: {e}")
-            models[model_name] = None
             return None
 
 for config in model_configs:
-    load_model(config)
+    model = load_model(config)
+    if model is None:
+        print(f"Failed to load model {config['name']}. Exiting.")
+        exit(1)
 
 
 class ChatRequest(BaseModel):
@@ -71,7 +61,7 @@ def remove_duplicates(text):
 def generate_model_response(model, inputs):
     try:
         if model is None:
-            return ""
+            return "Model loading failed."
         response = model(inputs)
         return remove_duplicates(response['choices'][0]['text'])
     except Exception as e:
@@ -112,7 +102,7 @@ async def process_message(message, history):
         response = requests.post(f"http://localhost:{port}/generate", json={"message": message}).json()
         formatted_response = response["response"]
         history.append((message, formatted_response))
-        return history, ""
+        return history, history
     except requests.exceptions.RequestException as e:
         return history, f"Error communicating with the backend: {e}"
 
@@ -122,7 +112,7 @@ iface = gr.Interface(
         gr.Textbox(lines=2, placeholder="Enter your message here..."),
         gr.State([])
     ],
-    outputs=[gr.Chatbot(), gr.Textbox(visible=False)],
+    outputs=[gr.Chatbot(), gr.State([])],
     title="Multi-Model LLM API",
     description="Enter a message and get responses from multiple LLMs.",
 )