diff --git a/nxontology_ml/text_embeddings/embeddings_model.py b/nxontology_ml/text_embeddings/embeddings_model.py
index 21b6d4f..b447963 100644
--- a/nxontology_ml/text_embeddings/embeddings_model.py
+++ b/nxontology_ml/text_embeddings/embeddings_model.py
@@ -119,7 +119,7 @@ def from_pretrained(
         """
         # FIXME: should we add truncation of input??
         cache_filename = (cache_path or _cache_path(pretrained_model_name)).as_posix()
-        logging.debug(f"Caching embeddings into: {cache_filename}")
+        logging.info(f"Caching embeddings into: {cache_filename}")
         return cls(
             lazy_model=lazy_model or _LazyAutoModel(pretrained_model_name),
             pooler_attr=_model_poolers[pretrained_model_name],