diff --git a/src/celeste/namespaces/domains.py b/src/celeste/namespaces/domains.py
index cd3265a6..7e2a4311 100644
--- a/src/celeste/namespaces/domains.py
+++ b/src/celeste/namespaces/domains.py
@@ -8,7 +8,7 @@
 
 from pydantic import SecretStr
 
-from celeste import create_client
+from celeste import Authentication, create_client
 from celeste.artifacts import ImageArtifact
 from celeste.core import Modality, Operation, Provider
 from celeste.modalities.audio.io import AudioOutput
@@ -38,6 +38,7 @@ def generate(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         base_url: str | None = None,
         extra_body: dict[str, Any] | None = None,
         **params: Unpack[TextParameters],
@@ -49,6 +50,7 @@ def generate(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.stream.generate(
             prompt,
@@ -70,6 +72,7 @@ def generate(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         base_url: str | None = None,
         extra_body: dict[str, Any] | None = None,
         **params: Unpack[TextParameters],
@@ -81,6 +84,7 @@ def generate(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.stream.generate(
             prompt,
@@ -102,6 +106,7 @@ def generate(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         base_url: str | None = None,
         extra_body: dict[str, Any] | None = None,
         **params: Unpack[TextParameters],
@@ -113,6 +118,7 @@ def generate(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.generate(
             prompt,
@@ -129,6 +135,7 @@ def embed(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[EmbeddingsParameters],
     ) -> EmbeddingsOutput:
         """Blocking embeddings generation."""
@@ -138,6 +145,7 @@ def embed(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.embed(text, **params)
 
@@ -161,6 +169,7 @@ async def generate(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         base_url: str | None = None,
         extra_body: dict[str, Any] | None = None,
         **parameters: Unpack[TextParameters],
@@ -173,6 +182,7 @@ async def generate(
             model: Model ID to use (required).
             provider: Optional provider override.
             api_key: Optional API key override.
+            auth: Optional Authentication object (e.g., GoogleADC for Vertex AI).
             base_url: Optional base URL for OpenResponses providers.
             extra_body: Optional provider-specific request body fields.
             **parameters: Additional model parameters.
@@ -186,6 +196,7 @@ async def generate(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         inputs = TextInput(prompt=prompt, messages=messages)
         return await client._predict(
@@ -199,6 +210,7 @@ async def embed(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **parameters: Unpack[EmbeddingsParameters],
     ) -> EmbeddingsOutput:
         """Generate embeddings from text.
@@ -208,6 +220,7 @@ async def embed(
             model: Model ID to use (required).
             provider: Optional provider override.
             api_key: Optional API key override.
+            auth: Optional Authentication object (e.g., GoogleADC for Vertex AI).
             **parameters: Additional model parameters.
 
         Returns:
@@ -219,6 +232,7 @@ async def embed(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return await client.embed(text, **parameters)
 
@@ -243,6 +257,7 @@ def generate(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[ImageParameters],
     ) -> ImagesStream:
         """Sync streaming image generation."""
@@ -252,6 +267,7 @@ def generate(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.stream.generate(prompt, **params)
 
@@ -263,6 +279,7 @@ def edit(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[ImageParameters],
     ) -> ImagesStream:
         """Sync streaming image editing."""
@@ -272,6 +289,7 @@ def edit(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.stream.edit(image, prompt, **params)
 
@@ -284,6 +302,7 @@ def analyze(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[TextParameters],
     ) -> TextStream:
         """Sync streaming image analysis."""
@@ -293,6 +312,7 @@ def analyze(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.stream.analyze(
             prompt, messages=messages, image=image, **params
@@ -309,6 +329,7 @@ def generate(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[ImageParameters],
     ) -> ImagesStream:
         """Async streaming image generation."""
@@ -318,6 +339,7 @@ def generate(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.stream.generate(prompt, **params)
 
@@ -329,6 +351,7 @@ def edit(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[ImageParameters],
     ) -> ImagesStream:
         """Async streaming image editing."""
@@ -338,6 +361,7 @@ def edit(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.stream.edit(image, prompt, **params)
 
@@ -350,6 +374,7 @@ def analyze(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[TextParameters],
     ) -> TextStream:
         """Async streaming image analysis."""
@@ -359,6 +384,7 @@ def analyze(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.stream.analyze(prompt, messages=messages, image=image, **params)
 
@@ -373,6 +399,7 @@ def generate(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[ImageParameters],
     ) -> ImageOutput:
         """Blocking image generation."""
@@ -382,6 +409,7 @@ def generate(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.generate(prompt, **params)
 
@@ -393,6 +421,7 @@ def edit(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[ImageParameters],
     ) -> ImageOutput:
         """Blocking image editing."""
@@ -402,6 +431,7 @@ def edit(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.edit(image, prompt, **params)
 
@@ -414,6 +444,7 @@ def analyze(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[TextParameters],
     ) -> TextOutput:
         """Blocking image analysis."""
@@ -423,6 +454,7 @@ def analyze(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.analyze(prompt, messages=messages, image=image, **params)
 
@@ -445,6 +477,7 @@ async def generate(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **parameters: Unpack[ImageParameters],
     ) -> ImageOutput:
         """Generate images from a prompt.
@@ -454,6 +487,7 @@ async def generate(
             model: Model ID to use (required).
             provider: Optional provider override.
             api_key: Optional API key override.
+            auth: Optional Authentication object (e.g., GoogleADC for Vertex AI).
             **parameters: Additional model parameters.
 
         Returns:
@@ -465,6 +499,7 @@ async def generate(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return await client.generate(prompt, **parameters)
 
@@ -476,6 +511,7 @@ async def edit(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **parameters: Unpack[ImageParameters],
     ) -> ImageOutput:
         """Edit an image with a prompt.
@@ -486,6 +522,7 @@ async def edit(
             model: Model ID to use (required).
             provider: Optional provider override.
             api_key: Optional API key override.
+            auth: Optional Authentication object (e.g., GoogleADC for Vertex AI).
             **parameters: Additional model parameters.
 
         Returns:
@@ -497,6 +534,7 @@ async def edit(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return await client.edit(image, prompt, **parameters)
 
@@ -509,6 +547,7 @@ async def analyze(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **parameters: Unpack[TextParameters],
     ) -> TextOutput:
         """Analyze images and return text description.
@@ -520,6 +559,7 @@ async def analyze(
             model: Model ID to use (required).
             provider: Optional provider override.
             api_key: Optional API key override.
+            auth: Optional Authentication object (e.g., GoogleADC for Vertex AI).
             **parameters: Additional model parameters.
 
         Returns:
@@ -531,6 +571,7 @@ async def analyze(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return await client.analyze(
             prompt, messages=messages, image=image, **parameters
@@ -557,6 +598,7 @@ def speak(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[AudioParameters],
     ) -> AudioStream:
         """Sync streaming text-to-speech."""
@@ -566,6 +608,7 @@ def speak(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.stream.speak(text, **params)
 
@@ -578,6 +621,7 @@ def analyze(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[TextParameters],
     ) -> TextStream:
         """Sync streaming audio analysis."""
@@ -587,6 +631,7 @@ def analyze(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.stream.analyze(
             prompt, messages=messages, audio=audio, **params
@@ -603,6 +648,7 @@ def speak(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[AudioParameters],
     ) -> AudioStream:
         """Async streaming text-to-speech."""
@@ -612,6 +658,7 @@ def speak(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.stream.speak(text, **params)
 
@@ -624,6 +671,7 @@ def analyze(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[TextParameters],
     ) -> TextStream:
         """Async streaming audio analysis."""
@@ -633,6 +681,7 @@ def analyze(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.stream.analyze(prompt, messages=messages, audio=audio, **params)
 
@@ -647,6 +696,7 @@ def speak(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[AudioParameters],
     ) -> AudioOutput:
         """Blocking text-to-speech."""
@@ -656,6 +706,7 @@ def speak(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.speak(text, **params)
 
@@ -668,6 +719,7 @@ def analyze(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[TextParameters],
     ) -> TextOutput:
         """Blocking audio analysis."""
@@ -677,6 +729,7 @@ def analyze(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.analyze(prompt, messages=messages, audio=audio, **params)
 
@@ -699,6 +752,7 @@ async def speak(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **parameters: Unpack[AudioParameters],
     ) -> AudioOutput:
         """Convert text to speech.
@@ -708,6 +762,7 @@ async def speak(
             model: Model ID to use (required).
             provider: Optional provider override.
             api_key: Optional API key override.
+            auth: Optional Authentication object (e.g., GoogleADC for Vertex AI).
             **parameters: Additional model parameters (e.g., voice).
 
         Returns:
@@ -719,6 +774,7 @@ async def speak(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return await client.speak(text, **parameters)
 
@@ -731,6 +787,7 @@ async def analyze(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **parameters: Unpack[TextParameters],
     ) -> TextOutput:
         """Analyze audio and return text transcription/description.
@@ -742,6 +799,7 @@ async def analyze(
             model: Model ID to use (required).
             provider: Optional provider override.
             api_key: Optional API key override.
+            auth: Optional Authentication object (e.g., GoogleADC for Vertex AI).
             **parameters: Additional model parameters.
 
         Returns:
@@ -753,6 +811,7 @@ async def analyze(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return await client.analyze(
             prompt, messages=messages, audio=audio, **parameters
@@ -781,6 +840,7 @@ def analyze(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[TextParameters],
     ) -> TextStream:
         """Sync streaming video analysis."""
@@ -790,6 +850,7 @@ def analyze(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.stream.analyze(
             prompt, messages=messages, video=video, **params
@@ -808,6 +869,7 @@ def analyze(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[TextParameters],
     ) -> TextStream:
         """Async streaming video analysis."""
@@ -817,6 +879,7 @@ def analyze(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.stream.analyze(prompt, messages=messages, video=video, **params)
 
@@ -831,6 +894,7 @@ def generate(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[VideoParameters],
     ) -> VideoOutput:
         """Blocking video generation."""
@@ -840,6 +904,7 @@ def generate(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.generate(prompt, **params)
 
@@ -852,6 +917,7 @@ def analyze(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **params: Unpack[TextParameters],
     ) -> TextOutput:
         """Blocking video analysis."""
@@ -861,6 +927,7 @@ def analyze(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return client.sync.analyze(prompt, messages=messages, video=video, **params)
 
@@ -883,6 +950,7 @@ async def generate(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **parameters: Unpack[VideoParameters],
     ) -> VideoOutput:
         """Generate video from a prompt.
@@ -892,6 +960,7 @@ async def generate(
             model: Model ID to use (required).
             provider: Optional provider override.
             api_key: Optional API key override.
+            auth: Optional Authentication object (e.g., GoogleADC for Vertex AI).
             **parameters: Additional model parameters.
 
         Returns:
@@ -903,6 +972,7 @@ async def generate(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return await client.generate(prompt, **parameters)
 
@@ -915,6 +985,7 @@ async def analyze(
         model: str,
         provider: Provider | None = None,
         api_key: str | SecretStr | None = None,
+        auth: Authentication | None = None,
         **parameters: Unpack[TextParameters],
     ) -> TextOutput:
         """Analyze video and return text description.
@@ -926,6 +997,7 @@ async def analyze(
             model: Model ID to use (required).
             provider: Optional provider override.
             api_key: Optional API key override.
+            auth: Optional Authentication object (e.g., GoogleADC for Vertex AI).
             **parameters: Additional model parameters.
 
         Returns:
@@ -937,6 +1009,7 @@ async def analyze(
             model=model,
             provider=provider,
             api_key=api_key,
+            auth=auth,
         )
         return await client.analyze(
             prompt, messages=messages, video=video, **parameters