diff --git a/cogs/commands.py b/cogs/commands.py
index f28c4bf..093d5a5 100644
--- a/cogs/commands.py
+++ b/cogs/commands.py
@@ -301,6 +301,11 @@ class Commands(discord.Cog, name="Commands"):
     @discord.option(
         name="prompt", description="The prompt to send to GPT3", required=True
     )
+    @discord.option(
+        name="ephemeral", 
+        description="Will only be visible to you", 
+        required=False
+    )
     @discord.option(
         name="temperature",
         description="Higher values means the model will take more risks",
@@ -334,13 +339,14 @@ class Commands(discord.Cog, name="Commands"):
         self,
         ctx: discord.ApplicationContext,
         prompt: str,
+        ephemeral: bool,
         temperature: float,
         top_p: float,
         frequency_penalty: float,
         presence_penalty: float,
     ):
         await self.converser_cog.ask_command(
-            ctx, prompt, temperature, top_p, frequency_penalty, presence_penalty
+            ctx, prompt, ephemeral, temperature, top_p, frequency_penalty, presence_penalty
         )
 
     @add_to_group("gpt")
@@ -360,6 +366,11 @@ class Commands(discord.Cog, name="Commands"):
         required=False,
         default="",
     )
+    @discord.option(
+        name="ephemeral", 
+        description="Will only be visible to you", 
+        required=False
+    )
     @discord.option(
         name="temperature",
         description="Higher values means the model will take more risks",
@@ -385,12 +396,13 @@ class Commands(discord.Cog, name="Commands"):
         ctx: discord.ApplicationContext,
         instruction: str,
         text: str,
+        ephemeral: bool,
         temperature: float,
         top_p: float,
         codex: bool,
     ):
         await self.converser_cog.edit_command(
-            ctx, instruction, text, temperature, top_p, codex
+            ctx, instruction, text, ephemeral, temperature, top_p, codex
         )
 
     @add_to_group("gpt")
diff --git a/cogs/text_service_cog.py b/cogs/text_service_cog.py
index b79d36a..2e01125 100644
--- a/cogs/text_service_cog.py
+++ b/cogs/text_service_cog.py
@@ -693,6 +693,7 @@ class GPT3ComCon(discord.Cog, name="GPT3ComCon"):
         self,
         ctx: discord.ApplicationContext,
         prompt: str,
+        ephemeral: bool,
         temperature: float,
         top_p: float,
         frequency_penalty: float,
@@ -720,7 +721,7 @@ class GPT3ComCon(discord.Cog, name="GPT3ComCon"):
             if not user_api_key:
                 return
 
-        await ctx.defer()
+        await ctx.defer(ephemeral=ephemeral)
 
         overrides = Override(temperature, top_p, frequency_penalty, presence_penalty)
 
@@ -741,6 +742,7 @@ class GPT3ComCon(discord.Cog, name="GPT3ComCon"):
         ctx: discord.ApplicationContext,
         instruction: str,
         text: str,
+        ephemeral: bool,
         temperature: float,
         top_p: float,
         codex: bool,
@@ -766,7 +768,7 @@ class GPT3ComCon(discord.Cog, name="GPT3ComCon"):
             if not user_api_key:
                 return
 
-        await ctx.defer()
+        await ctx.defer(ephemeral=ephemeral)
 
         overrides = Override(temperature, top_p, 0, 0)
 
diff --git a/models/openai_model.py b/models/openai_model.py
index 24ebcc6..338df39 100644
--- a/models/openai_model.py
+++ b/models/openai_model.py
@@ -584,7 +584,7 @@ class Model:
     def backoff_handler_request(details):
         print(
             f"Backing off {details['wait']:0.1f} seconds after {details['tries']} tries calling function {details['target']} | "
-            f"{details['exception'].args}"
+            f"{details['exception'].args[0]}"
         )
 
     async def valid_text_request(self, response):
@@ -657,7 +657,7 @@ class Model:
         )
         print(f"Overrides -> temp:{temp_override}, top_p:{top_p_override}")
 
-        async with aiohttp.ClientSession(raise_for_status=True) as session:
+        async with aiohttp.ClientSession(raise_for_status=False) as session:
             payload = {
                 "model": Models.EDIT if codex is False else Models.CODE_EDIT,
                 "input": "" if text is None else text,
@@ -724,7 +724,7 @@ class Model:
 
         tokens = self.usage_service.count_tokens(summary_request_text)
 
-        async with aiohttp.ClientSession(raise_for_status=True) as session:
+        async with aiohttp.ClientSession(raise_for_status=False) as session:
             payload = {
                 "model": Models.DAVINCI,
                 "prompt": summary_request_text,