diff --git a/llama_stack/providers/utils/inference/embedding_mixin.py b/llama_stack/providers/utils/inference/embedding_mixin.py index 2a16e7f40..8b14c7502 100644 --- a/llama_stack/providers/utils/inference/embedding_mixin.py +++ b/llama_stack/providers/utils/inference/embedding_mixin.py @@ -43,7 +43,7 @@ class SentenceTransformerEmbeddingMixin: ) return EmbeddingsResponse(embeddings=embeddings) - def _load_sentence_transformer_model(self, model: str) -> SentenceTransformer: + def _load_sentence_transformer_model(self, model: str) -> "SentenceTransformer": global EMBEDDING_MODELS loaded_model = EMBEDDING_MODELS.get(model)