Update app.py

app.py

@@ -6,6 +6,7 @@ from fastapi.responses import StreamingResponse
 import uvicorn
 from dotenv import load_dotenv
 import os
+import re
 
 # Load environment variables from .env file
 load_dotenv()
@@ -27,7 +28,6 @@ class v1:
     ):
         """
         Initializes the v1 AI API with given parameters.
-
         Args:
             model (str, optional): The AI model to use for text generation. Defaults to "claude". 
                                     Options: "llama", "claude".
@@ -78,6 +78,7 @@ class v1:
                 f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
             )
 
+        buffer = ""
         for line in response.iter_lines(decode_unicode=True):
             if line:
                 if line.startswith("data: "):
@@ -90,14 +91,29 @@ class v1:
                                (self.model == "claude" and model == 'OPENROUTER_CLAUDE'):
                                 content = data['chunk']['content']
                                 if content:
-                                    yield content
+                                    buffer += content
+                                    # Check if we have a complete line or paragraph
+                                    lines = buffer.split('\n')
+                                    if len(lines) > 1:
+                                        for complete_line in lines[:-1]:
+                                            yield self.format_text(complete_line) + '\n'
+                                        buffer = lines[-1]
                     except KeyError:
                         pass
                     except json.JSONDecodeError:
                         pass
 
+        # Yield any remaining content in the buffer
+        if buffer:
+            yield self.format_text(buffer)
+
         yield "[DONE]"
 
+    def format_text(self, text: str) -> str:
+        # Convert *text* to <i>text</i> for italic
+        text = re.sub(r'\*(.*?)\*', r'<i>\1</i>', text)
+        return text
+
     def chat(self, prompt: str) -> Generator[str, None, None]:
         """Stream responses as string chunks"""
         return self.ask(prompt)