Reimplement replicate inference. Allow model deletion. Implement repetition penalty. Fixes in stream reading.

backend/src/app.py

@@ -253,6 +253,8 @@ class App(ChatMixin):
     async def get_model(self, model_id):
         return await db.Model.get(id=model_id)
 
+    # Serves for creating and updating models.
+    # Deletes the model by the id if all other parameters are empty.
     async def save_model(self, model_id, kind, name, value):
         if model_id is None:
             model = await db.Model.create(
@@ -261,12 +263,17 @@ class App(ChatMixin):
                 value=value,
             )
         else:
-            model = await db.Model.update_or_create(
-                id=model_id, 
-                name=name, 
-                kind=kind, 
-                value=value,
-            )
+            if value is None and name is None and kind is None:
+                await db.Model.filter(id=model_id).delete()
+                print(f"Deleted model: {model_id}")
+                return None
+            else:
+                model = await db.Model.update_or_create(
+                    id=model_id, 
+                    name=name, 
+                    kind=kind, 
+                    value=value,
+                )
         print(f"Saved model: {model.id} - {name}")
         return model