diff --git a/reme/core/embedding/base_embedding_model.py b/reme/core/embedding/base_embedding_model.py
index b9871a34..e40526d0 100644
--- a/reme/core/embedding/base_embedding_model.py
+++ b/reme/core/embedding/base_embedding_model.py
@@ -6,7 +6,6 @@ Defines the abstract base class and standard API for all embedding model impleme
 import asyncio
 import hashlib
 import json
-import os
 import time
 from abc import ABC
 from collections import OrderedDict
@@ -56,8 +55,8 @@ class BaseEmbeddingModel(ABC):
             enable_cache: Whether to enable embedding cache
             **kwargs: Additional model-specific parameters
         """
-        self._api_key: str | None = api_key
-        self._base_url: str | None = base_url
+        self.api_key: str | None = api_key
+        self.base_url: str | None = base_url
         self.model_name = model_name
         self.dimensions = dimensions
         self.use_dimensions = use_dimensions
@@ -78,25 +77,6 @@ class BaseEmbeddingModel(ABC):
         self.cache_path: Path = Path(self.cache_dir)
         self.cache_path.mkdir(parents=True, exist_ok=True)
 
-    @property
-    def api_key(self) -> str | None:
-        """Get API key from environment variable."""
-        return (
-            os.getenv("REME_EMBEDDING_API_KEY")
-            or os.getenv("EMBEDDING_API_KEY")
-            or os.getenv("OPENAI_API_KEY")
-            or self._api_key
-        )
-
-    @property
-    def base_url(self) -> str | None:
-        """Get base URL from environment variable."""
-        return (
-            os.getenv("REME_EMBEDDING_BASE_URL")
-            or os.getenv("EMBEDDING_BASE_URL")
-            or self._base_url
-        )
-
     def _truncate_text(self, text: str) -> str:
         """Truncate text to max_input_length if it exceeds the limit."""
         if len(text) > self.max_input_length:
diff --git a/tests/test_file_store.py b/tests/test_file_store.py
index 847486ee..8c300b97 100644
--- a/tests/test_file_store.py
+++ b/tests/test_file_store.py
@@ -243,7 +243,7 @@ def create_file_store(store_type: str) -> BaseFileStore:
     """
     config = TestConfig()
 
-    # Initialize embedding model (api_key/base_url from env: EMBEDDING_* or OPENAI_API_KEY)
+    # api_key/base_url from env (same as Application.embedding_api_key in application.py)
     embedding_api_key = os.environ.get("EMBEDDING_API_KEY") or os.environ.get("OPENAI_API_KEY")
     embedding_base_url = os.environ.get("EMBEDDING_BASE_URL") or os.environ.get("OPENAI_BASE_URL")
     # use_dimensions=True so API returns 64-dim, matching collection/table schema for all backends
diff --git a/tests/test_vector_store.py b/tests/test_vector_store.py
index 7861e7bf..d1e2a9c5 100644
--- a/tests/test_vector_store.py
+++ b/tests/test_vector_store.py
@@ -278,11 +278,16 @@ def create_vector_store(store_type: str, collection_name: str) -> BaseVectorStor
     """
     config = TestConfig()
 
-    # Initialize embedding model (use_dimensions=True so API output matches HNSW dim / vector column)
+    # api_key/base_url from env (Application layer does the same via embedding_api_key in application.py)
+    embedding_api_key = os.environ.get("EMBEDDING_API_KEY") or os.environ.get("OPENAI_API_KEY")
+    embedding_base_url = os.environ.get("EMBEDDING_BASE_URL") or os.environ.get("OPENAI_BASE_URL")
+    # use_dimensions=True so API output matches HNSW dim / vector column
     embedding_model = OpenAIEmbeddingModel(
         model_name=config.EMBEDDING_MODEL_NAME,
         dimensions=config.EMBEDDING_DIMENSIONS,
         use_dimensions=True,
+        api_key=embedding_api_key or None,
+        base_url=embedding_base_url or None,
     )
 
     if store_type == "local":