huangchengtao
diff --git a/‎llama_cpp/server/model.py
Copy file name to clipboardExpand all lines: llama_cpp/server/model.py
+1Lines changed: 1 addition & 0 deletions b/‎llama_cpp/server/model.py
Copy file name to clipboardExpand all lines: llama_cpp/server/model.py
+1Lines changed: 1 addition & 0 deletions
@@ -223,6 +223,7 @@ def load_llama_from_model_settings(settings: ModelSettings) -> llama_cpp.Llama:
             **kwargs,
             # Model Params
             n_gpu_layers=settings.n_gpu_layers,
+            split_mode=settings.split_mode,
             main_gpu=settings.main_gpu,
             tensor_split=settings.tensor_split,
             vocab_only=settings.vocab_only,