deprecate: drop fallback url to a google bucket

fastembed/common/model_management.py

@@ -5,6 +5,7 @@
 import shutil
 import tarfile
 import tempfile
+import warnings
 import contextlib
 from copy import deepcopy
 from pathlib import Path, PureWindowsPath
@@ -374,6 +375,18 @@ def retrieve_model_gcs(
 
         # check if the model_dir and the model files are both present for macOS
         if model_dir.exists() and len(list(model_dir.glob("*"))) > 0:
+            if deprecated_tar_struct:
+                # No built-in model is served from the bucket anymore, only copies of it remain.
+                # stacklevel points at the caller of TextEmbedding(...), the path that gets here.
+                warnings.warn(
+                    f"Loading {model_name} from {model_dir}, a copy downloaded from Google Cloud "
+                    "Storage by an older fastembed version. Support for such copies is deprecated "
+                    "and will be removed in a future release. To switch to Hugging Face, load the "
+                    "model once with network access (without `local_files_only=True` or "
+                    f"`HF_HUB_OFFLINE=1`), then delete {model_dir}.",
+                    FutureWarning,
+                    stacklevel=5,
+                )
             return model_dir
 
         if local_files_only:
@@ -440,7 +453,6 @@ def download_model(cls, model: T, cache_dir: str, retries: int = 3, **kwargs: An
                     "description": "Base English model, v1.5",
                     "size_in_GB": 0.44,
                     "sources": {
-                        "url": "https://storage.googleapis.com/qdrant-fastembed/fast-bge-base-en-v1.5.tar.gz",
                         "hf": "qdrant/bge-base-en-v1.5-onnx-q",
                     }
                 }

fastembed/text/onnx_embedding.py

@@ -19,7 +19,6 @@
         size_in_GB=0.42,
         sources=ModelSource(
             hf="Qdrant/fast-bge-base-en",
-            url="https://storage.googleapis.com/qdrant-fastembed/fast-bge-base-en.tar.gz",
             _deprecated_tar_struct=True,
         ),
         model_file="model_optimized.onnx",
@@ -35,7 +34,6 @@
         size_in_GB=0.21,
         sources=ModelSource(
             hf="Qdrant/bge-base-en-v1.5-onnx-Q",
-            url="https://storage.googleapis.com/qdrant-fastembed/fast-bge-base-en-v1.5.tar.gz",
             _deprecated_tar_struct=True,
         ),
         model_file="model_optimized.onnx",
@@ -63,7 +61,6 @@
         size_in_GB=0.13,
         sources=ModelSource(
             hf="Qdrant/bge-small-en",
-            url="https://storage.googleapis.com/qdrant-fastembed/BAAI-bge-small-en.tar.gz",
             _deprecated_tar_struct=True,
         ),
         model_file="model_optimized.onnx",
@@ -91,7 +88,6 @@
         size_in_GB=0.09,
         sources=ModelSource(
             hf="Qdrant/bge-small-zh-v1.5",
-            url="https://storage.googleapis.com/qdrant-fastembed/fast-bge-small-zh-v1.5.tar.gz",
             _deprecated_tar_struct=True,
         ),
         model_file="model_optimized.onnx",

fastembed/text/pooled_embedding.py

@@ -81,7 +81,6 @@
         size_in_GB=2.24,
         sources=ModelSource(
             hf="qdrant/multilingual-e5-large-onnx",
-            url="https://storage.googleapis.com/qdrant-fastembed/fast-multilingual-e5-large.tar.gz",
             _deprecated_tar_struct=True,
         ),
         model_file="model.onnx",

fastembed/text/pooled_normalized_embedding.py

@@ -19,7 +19,6 @@
         license="apache-2.0",
         size_in_GB=0.09,
         sources=ModelSource(
-            url="https://storage.googleapis.com/qdrant-fastembed/sentence-transformers-all-MiniLM-L6-v2.tar.gz",
             hf="qdrant/all-MiniLM-L6-v2-onnx",
             _deprecated_tar_struct=True,
         ),