diff --git a/whisperx/diarize.py b/whisperx/diarize.py
index e50dc0f..d59a847 100644
--- a/whisperx/diarize.py
+++ b/whisperx/diarize.py
@@ -6,10 +6,11 @@ import torch
 
 from .audio import load_audio, SAMPLE_RATE
 
+
 class DiarizationPipeline:
     def __init__(
         self,
-        model_name="pyannote/speaker-diarization@2.1",
+        model_name="pyannote/speaker-diarization-3.0",
         use_auth_token=None,
         device: Optional[Union[str, torch.device]] = "cpu",
     ):