diff --git a/CHANGELOG.md b/CHANGELOG.md
index 83320ed91c915..3b68072555958 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -4,6 +4,7 @@
 
 ### Bug Fixes / Nits
 - tune prompt to get rid of KeyError in SubQ engine (#7039)
+- Fix validation of Azure OpenAI keys (#7042)
 
 ## [0.7.12] - 2023-07-25
 
diff --git a/llama_index/embeddings/openai.py b/llama_index/embeddings/openai.py
index c033b496894b2..2c7a11bbe052d 100644
--- a/llama_index/embeddings/openai.py
+++ b/llama_index/embeddings/openai.py
@@ -245,7 +245,9 @@ def __init__(
         # Validate that either the openai.api_key property
         # or OPENAI_API_KEY env variable are set to a valid key
         # Raises ValueError if missing or doesn't match valid format
-        validate_openai_api_key(kwargs.get("api_key", None))
+        validate_openai_api_key(
+            kwargs.get("api_key", None), kwargs.get("api_type", None)
+        )
 
         """Init params."""
         super().__init__(embed_batch_size, tokenizer, callback_manager)
diff --git a/llama_index/llms/openai.py b/llama_index/llms/openai.py
index de6e20d3e69a2..30b47048272a6 100644
--- a/llama_index/llms/openai.py
+++ b/llama_index/llms/openai.py
@@ -43,7 +43,9 @@ class OpenAI(LLM, BaseModel):
     max_retries: int = 10
 
     def __init__(self, *args: Any, **kwargs: Any) -> None:
-        validate_openai_api_key(kwargs.get("api_key", None))
+        validate_openai_api_key(
+            kwargs.get("api_key", None), kwargs.get("api_type", None)
+        )
         super().__init__(*args, **kwargs)
 
     @property
diff --git a/llama_index/llms/openai_utils.py b/llama_index/llms/openai_utils.py
index 2594be2591b9f..fbced5e1f1935 100644
--- a/llama_index/llms/openai_utils.py
+++ b/llama_index/llms/openai_utils.py
@@ -256,9 +256,17 @@ def to_openai_function(pydantic_class: Type[BaseModel]) -> Dict[str, Any]:
     }
 
 
-def validate_openai_api_key(api_key: Optional[str] = None) -> None:
+def validate_openai_api_key(
+    api_key: Optional[str] = None, api_type: Optional[str] = None
+) -> None:
     openai_api_key = api_key or os.environ.get("OPENAI_API_KEY", "") or openai.api_key
+    openai_api_type = (
+        api_type or os.environ.get("OPENAI_API_TYPE", "") or openai.api_type
+    )
+
     if not openai_api_key:
         raise ValueError(MISSING_API_KEY_ERROR_MESSAGE)
-    elif not OPENAI_API_KEY_FORMAT.search(openai_api_key):
+    elif openai_api_type == "open_ai" and not OPENAI_API_KEY_FORMAT.search(
+        openai_api_key
+    ):
         raise ValueError(INVALID_API_KEY_ERROR_MESSAGE)
diff --git a/tests/conftest.py b/tests/conftest.py
index 2c4d292c680f3..24e0158631eb9 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -93,8 +93,8 @@ def mock_openai_credentials() -> None:
 
 class CachedOpenAIApiKeys:
     """
-    Saves the users' OpenAI API key either in the environment variable
-    or set to the library itself.
+    Saves the users' OpenAI API key and OpenAI API type either in
+    the environment variable or set to the library itself.
     This allows us to run tests by setting it without plowing over
     the local environment.
     """
@@ -104,17 +104,25 @@ def __init__(
         set_env_key_to: Optional[str] = "",
         set_library_key_to: Optional[str] = None,
         set_fake_key: bool = False,
+        set_env_type_to: Optional[str] = "",
+        set_library_type_to: str = "open_ai",  # default value in openai package
     ):
         self.set_env_key_to = set_env_key_to
         self.set_library_key_to = set_library_key_to
         self.set_fake_key = set_fake_key
+        self.set_env_type_to = set_env_type_to
+        self.set_library_type_to = set_library_type_to
 
     def __enter__(self) -> None:
         self.api_env_variable_was = os.environ.get("OPENAI_API_KEY", "")
+        self.api_env_type_was = os.environ.get("OPENAI_API_TYPE", "")
         self.openai_api_key_was = openai.api_key
+        self.openai_api_type_was = openai.api_type
 
         os.environ["OPENAI_API_KEY"] = str(self.set_env_key_to)
+        os.environ["OPENAI_API_TYPE"] = str(self.set_env_type_to)
         openai.api_key = self.set_library_key_to
+        openai.api_type = self.set_library_type_to
 
         if self.set_fake_key:
             openai.api_key = "sk-" + "a" * 48
@@ -122,4 +130,6 @@ def __enter__(self) -> None:
     # No matter what, set the environment variable back to what it was
     def __exit__(self, *exc: Any) -> None:
         os.environ["OPENAI_API_KEY"] = str(self.api_env_variable_was)
+        os.environ["OPENAI_API_TYPE"] = str(self.api_env_type_was)
         openai.api_key = self.openai_api_key_was
+        openai.api_type = self.openai_api_type_was
diff --git a/tests/embeddings/test_base.py b/tests/embeddings/test_base.py
index 2fe279edd411d..099e6ac33faa0 100644
--- a/tests/embeddings/test_base.py
+++ b/tests/embeddings/test_base.py
@@ -128,8 +128,18 @@ def test_validates_api_key_format_from_env() -> None:
         with pytest.raises(ValueError, match="Invalid OpenAI API key."):
             OpenAIEmbedding()
 
+    with CachedOpenAIApiKeys(
+        set_env_key_to="api-hf47930g732gf372", set_env_type_to="azure"
+    ):
+        assert OpenAIEmbedding()
+
 
 def test_validates_api_key_format_in_library() -> None:
     with CachedOpenAIApiKeys(set_library_key_to="api-hf47930g732gf372"):
         with pytest.raises(ValueError, match="Invalid OpenAI API key."):
             OpenAIEmbedding()
+
+    with CachedOpenAIApiKeys(
+        set_library_key_to="api-hf47930g732gf372", set_library_type_to="azure"
+    ):
+        assert OpenAIEmbedding()