diff --git a/python/packages/openai/agent_framework_openai/_chat_client.py b/python/packages/openai/agent_framework_openai/_chat_client.py
index 942b0cc57bd..57275a26c8b 100644
--- a/python/packages/openai/agent_framework_openai/_chat_client.py
+++ b/python/packages/openai/agent_framework_openai/_chat_client.py
@@ -117,6 +117,7 @@
 from ._shared import (
     AzureTokenProvider,
     _attach_prompt_cache_breakpoint,  # pyright: ignore[reportPrivateUsage]
+    _is_mp3_media_type,  # pyright: ignore[reportPrivateUsage]
     load_openai_service_settings,
     maybe_append_azure_endpoint_guidance,
 )
@@ -2150,7 +2151,7 @@ def _prepare_content_for_openai(
                 if content.has_top_level_media_type("audio"):
                     if content.media_type and "wav" in content.media_type:
                         format = "wav"
-                    elif content.media_type and "mp3" in content.media_type:
+                    elif _is_mp3_media_type(content.media_type):
                         format = "mp3"
                     else:
                         logger.warning("Unsupported audio media type: %s", content.media_type)
diff --git a/python/packages/openai/agent_framework_openai/_chat_completion_client.py b/python/packages/openai/agent_framework_openai/_chat_completion_client.py
index a816a391259..ff88f65cd04 100644
--- a/python/packages/openai/agent_framework_openai/_chat_completion_client.py
+++ b/python/packages/openai/agent_framework_openai/_chat_completion_client.py
@@ -67,6 +67,7 @@
     PROMPT_CACHE_BREAKPOINT_KEY,
     AzureTokenProvider,
     _attach_prompt_cache_breakpoint,  # pyright: ignore[reportPrivateUsage]
+    _is_mp3_media_type,  # pyright: ignore[reportPrivateUsage]
     load_openai_service_settings,
     maybe_append_azure_endpoint_guidance,
 )
@@ -1315,7 +1316,7 @@ def _prepare_content_for_openai(self, content: Content) -> dict[str, Any]:
             case "data" | "uri" if content.has_top_level_media_type("audio"):
                 if content.media_type and "wav" in content.media_type:
                     audio_format = "wav"
-                elif content.media_type and "mp3" in content.media_type:
+                elif _is_mp3_media_type(content.media_type):
                     audio_format = "mp3"
                 else:
                     logger.debug("Unsupported audio media type: %s", content.media_type)
diff --git a/python/packages/openai/agent_framework_openai/_shared.py b/python/packages/openai/agent_framework_openai/_shared.py
index 2541f20b512..77e7dac867b 100644
--- a/python/packages/openai/agent_framework_openai/_shared.py
+++ b/python/packages/openai/agent_framework_openai/_shared.py
@@ -56,6 +56,29 @@
 PROMPT_CACHE_BREAKPOINT_KEY = "prompt_cache_breakpoint"
 
 
+MP3_MEDIA_TYPES = frozenset({"audio/mp3", "audio/mpeg", "audio/x-mpeg", "audio/mpeg3", "audio/x-mpeg-3"})
+
+
+def _is_mp3_media_type(  # pyright: ignore[reportUnusedFunction]
+    media_type: str | None,
+) -> bool:
+    """Check whether an audio media type names MP3 audio.
+
+    The media type is stripped of parameters and case-folded before it is compared against
+    the registered MP3 aliases, so that sibling MPEG subtypes such as ``audio/mpegurl`` or
+    ``audio/mpeg4-generic`` are not mistaken for MP3.
+
+    Args:
+        media_type: The media type of the audio content, if any.
+
+    Returns:
+        True if the media type is an alias of MP3 audio, False otherwise.
+    """
+    if not media_type:
+        return False
+    return media_type.partition(";")[0].strip().lower() in MP3_MEDIA_TYPES
+
+
 def _attach_prompt_cache_breakpoint(  # pyright: ignore[reportUnusedFunction]
     part: dict[str, Any], content: Content
 ) -> dict[str, Any]:
diff --git a/python/packages/openai/tests/openai/test_openai_chat_client.py b/python/packages/openai/tests/openai/test_openai_chat_client.py
index dc00749b337..c0a822dabf7 100644
--- a/python/packages/openai/tests/openai/test_openai_chat_client.py
+++ b/python/packages/openai/tests/openai/test_openai_chat_client.py
@@ -7717,6 +7717,32 @@ def test_prepare_content_for_openai_audio_content() -> None:
     assert result["type"] == "input_audio"
     assert result["input_audio"]["format"] == "mp3"
 
+    # Test MP3 audio content with the registered media type
+    mpeg_content = Content.from_uri(uri="data:audio/mpeg;base64,ghi789", media_type="audio/mpeg")
+    result = client._prepare_content_for_openai("user", mpeg_content)
+    assert result["type"] == "input_audio"
+    assert result["input_audio"]["data"] == "data:audio/mpeg;base64,ghi789"
+    assert result["input_audio"]["format"] == "mp3"
+
+    # Test MP3 aliases with media type parameters and mixed case
+    parameterized_content = Content.from_uri(
+        uri="data:audio/mpeg;base64,jkl012",
+        media_type="Audio/MPEG; codecs=mpeg-3",
+    )
+    result = client._prepare_content_for_openai("user", parameterized_content)
+    assert result["type"] == "input_audio"
+    assert result["input_audio"]["format"] == "mp3"
+
+
+def test_prepare_content_for_openai_non_mp3_mpeg_audio_is_unsupported() -> None:
+    """Test _prepare_content_for_openai omits MPEG audio that is not MP3."""
+    client = OpenAIChatClient(model="test-model", api_key="test-key")
+
+    for media_type in ("audio/mpegurl", "audio/mpeg4-generic"):
+        content = Content.from_uri(uri=f"data:{media_type};base64,abc123", media_type=media_type)
+
+        assert client._prepare_content_for_openai("user", content) == {}
+
 
 def test_prepare_content_for_openai_unsupported_content() -> None:
     """Test _prepare_content_for_openai with unsupported content types."""
diff --git a/python/packages/openai/tests/openai/test_openai_chat_completion_client.py b/python/packages/openai/tests/openai/test_openai_chat_completion_client.py
index acdc233f4c2..e01c3d92d45 100644
--- a/python/packages/openai/tests/openai/test_openai_chat_completion_client.py
+++ b/python/packages/openai/tests/openai/test_openai_chat_completion_client.py
@@ -602,11 +602,44 @@ def test_prepare_content_for_openai_data_content_image(
     assert result["input_audio"]["data"] == "//uQAAAAWGluZwAAAA8AAAACAAACcQ=="
     assert result["input_audio"]["format"] == "mp3"
 
+    # Test DataContent with MP3 audio using the registered media type
+    mpeg_data_content = Content.from_uri(
+        uri="data:audio/mpeg;base64,//uQAAAAWGluZwAAAA8AAAACAAACcQ==",
+        media_type="audio/mpeg",
+    )
+
+    result = client._prepare_content_for_openai(mpeg_data_content)  # type: ignore
+
+    assert result["type"] == "input_audio"
+    assert result["input_audio"]["data"] == "//uQAAAAWGluZwAAAA8AAAACAAACcQ=="
+    assert result["input_audio"]["format"] == "mp3"
+
+    # Test MP3 aliases with media type parameters and mixed case
+    parameterized_content = Content.from_uri(
+        uri="data:audio/mpeg;base64,//uQAAAAWGluZwAAAA8AAAACAAACcQ==",
+        media_type="Audio/MPEG; codecs=mpeg-3",
+    )
+
+    result = client._prepare_content_for_openai(parameterized_content)  # type: ignore
+
+    assert result["type"] == "input_audio"
+    assert result["input_audio"]["format"] == "mp3"
+
     unsupported_audio = Content.from_uri(uri="data:audio/ogg;base64,abc123", media_type="audio/ogg")
 
     assert client._prepare_content_for_openai(unsupported_audio) == {}  # type: ignore
 
 
+def test_prepare_content_for_openai_non_mp3_mpeg_audio_is_unsupported() -> None:
+    """Test _prepare_content_for_openai omits MPEG audio that is not MP3."""
+    client = OpenAIChatCompletionClient(model="test-model", api_key="test-key")
+
+    for media_type in ("audio/mpegurl", "audio/mpeg4-generic"):
+        content = Content.from_uri(uri=f"data:{media_type};base64,abc123", media_type=media_type)
+
+        assert client._prepare_content_for_openai(content) == {}  # type: ignore
+
+
 def test_prepare_message_for_openai_omits_unsupported_content() -> None:
     client = OpenAIChatCompletionClient(model="test-model", api_key="test-key")
     unsupported = Content.from_uri(uri="data:text/plain;base64,SGVsbG8=", media_type="text/plain")