diff --git a/studio/backend/core/inference/llama_cpp.py b/studio/backend/core/inference/llama_cpp.py
index 12b2495834..e793114f30 100644
--- a/studio/backend/core/inference/llama_cpp.py
+++ b/studio/backend/core/inference/llama_cpp.py
@@ -741,6 +741,7 @@ class LlamaCppBackend:
                 LlamaCppBackend._codec_mgr.unload()
                 LlamaCppBackend._codec_mgr = None
                 import torch
+
                 if torch.cuda.is_available():
                     torch.cuda.empty_cache()
             return True
diff --git a/studio/backend/routes/inference.py b/studio/backend/routes/inference.py
index 2d4e3a609b..b6f82bfe08 100644
--- a/studio/backend/routes/inference.py
+++ b/studio/backend/routes/inference.py
@@ -545,9 +545,12 @@ async def generate_audio(
     if llama_backend.is_loaded and getattr(llama_backend, "_is_audio", False):
         model_name = llama_backend.model_identifier
         gen = lambda: llama_backend.generate_audio_response(
-            text = text, audio_type = llama_backend._audio_type,
-            temperature = payload.temperature, top_p = payload.top_p,
-            top_k = payload.top_k, min_p = payload.min_p,
+            text = text,
+            audio_type = llama_backend._audio_type,
+            temperature = payload.temperature,
+            top_p = payload.top_p,
+            top_k = payload.top_k,
+            min_p = payload.min_p,
             max_new_tokens = payload.max_tokens or 2048,
             repetition_penalty = payload.repetition_penalty,
         )