diff --git a/backend/open_webui/routers/audio.py b/backend/open_webui/routers/audio.py
index 1599852f3c..17bb32f35c 100644
--- a/backend/open_webui/routers/audio.py
+++ b/backend/open_webui/routers/audio.py
@@ -1091,19 +1091,17 @@ async def transcribe(request: Request, file_path: str, metadata: Optional[dict]
                 detail=ERROR_MESSAGES.DEFAULT(e, 'Error processing audio file'),
             )
 
-    results = []
     try:
         tasks = [transcription_handler(request, chunk_path, metadata, user) for chunk_path in chunk_paths]
-        for coro in asyncio.as_completed(tasks):
-            try:
-                results.append(await coro)
-            except HTTPException:
-                raise
-            except Exception as transcribe_exc:
-                raise HTTPException(
-                    status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
-                    detail=f'Error transcribing chunk: {transcribe_exc}',
-                )
+        # gather keeps results in chunk order, unlike as_completed
+        results = await asyncio.gather(*tasks)
+    except HTTPException:
+        raise
+    except Exception as transcribe_exc:
+        raise HTTPException(
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
+            detail=f'Error transcribing chunk: {transcribe_exc}',
+        )
     finally:
         # Clean up only the temporary chunks, never the original file
         for chunk_path in chunk_paths: