docs(api): clarify prompt parameter support in audio transcriptions/translations

Author: stainless-app[bot]

.stats.yml

@@ -1,4 +1,4 @@
 configured_endpoints: 75
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/togetherai/togetherai-13728ec693ca29a36a3cfbe05df62ca69a8e404eab7f6db16759fc5304717eac.yml
-openapi_spec_hash: 5334a20f914487feace9c88045b89354
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/togetherai/togetherai-c120da21adf22dd3902475f6ba7410dd57d9d2995a232e676d59af4806d57d09.yml
+openapi_spec_hash: eff04023d0ad22f11bbf107e44fe699e
 config_hash: 6c214c91fad5ead4849be777fd9e8108

src/together/resources/audio/transcriptions.py

@@ -94,7 +94,10 @@ def create(
 
           model: Model to use for transcription
 
-          prompt: Optional text to bias decoding.
+          prompt: Optional text to bias decoding. Supported only on Whisper-family models (e.g.
+              `openai/whisper-large-v3`). Other STT models (e.g.
+              `nvidia/parakeet-tdt-0.6b-v3`) accept the field for API compatibility but ignore
+              it.
 
           response_format: The format of the response
 
@@ -217,7 +220,10 @@ async def create(
 
           model: Model to use for transcription
 
-          prompt: Optional text to bias decoding.
+          prompt: Optional text to bias decoding. Supported only on Whisper-family models (e.g.
+              `openai/whisper-large-v3`). Other STT models (e.g.
+              `nvidia/parakeet-tdt-0.6b-v3`) accept the field for API compatibility but ignore
+              it.
 
           response_format: The format of the response
 

src/together/resources/audio/translations.py

@@ -74,7 +74,10 @@ def create(
 
           model: Model to use for translation
 
-          prompt: Optional text to bias decoding.
+          prompt: Optional text to bias decoding. Supported only on Whisper-family models (e.g.
+              `openai/whisper-large-v3`). Other STT models (e.g.
+              `nvidia/parakeet-tdt-0.6b-v3`) accept the field for API compatibility but ignore
+              it.
 
           response_format: The format of the response
 
@@ -174,7 +177,10 @@ async def create(
 
           model: Model to use for translation
 
-          prompt: Optional text to bias decoding.
+          prompt: Optional text to bias decoding. Supported only on Whisper-family models (e.g.
+              `openai/whisper-large-v3`). Other STT models (e.g.
+              `nvidia/parakeet-tdt-0.6b-v3`) accept the field for API compatibility but ignore
+              it.
 
           response_format: The format of the response
 

src/together/types/audio/transcription_create_params.py

@@ -55,7 +55,12 @@ class TranscriptionCreateParams(TypedDict, total=False):
     """Model to use for transcription"""
 
     prompt: str
-    """Optional text to bias decoding."""
+    """Optional text to bias decoding.
+
+    Supported only on Whisper-family models (e.g. `openai/whisper-large-v3`). Other
+    STT models (e.g. `nvidia/parakeet-tdt-0.6b-v3`) accept the field for API
+    compatibility but ignore it.
+    """
 
     response_format: Literal["json", "verbose_json"]
     """The format of the response"""

src/together/types/audio/translation_create_params.py

@@ -27,7 +27,12 @@ class TranslationCreateParams(TypedDict, total=False):
     """Model to use for translation"""
 
     prompt: str
-    """Optional text to bias decoding."""
+    """Optional text to bias decoding.
+
+    Supported only on Whisper-family models (e.g. `openai/whisper-large-v3`). Other
+    STT models (e.g. `nvidia/parakeet-tdt-0.6b-v3`) accept the field for API
+    compatibility but ignore it.
+    """
 
     response_format: Literal["json", "verbose_json"]
     """The format of the response"""