From 11e45ad95361eb37c00f9cce799bf544fd9990a2 Mon Sep 17 00:00:00 2001
From: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
Date: Sat, 5 Sep 2026 23:00:01 -0700
Subject: [PATCH] fix(vertex_ai): mark the Lyria 3 catalog entries text-only

`vertex_ai/lyria-3-clip-preview` and `vertex_ai/lyria-3-pro-preview` were
registered with `supports_vision`, `supports_image_input`, and an `image`
modality, which contradicts their `gemini/lyria-3-*` siblings and makes
/model/info advertise image input on text-to-music models.
---
 .../model_prices_and_context_window_backup.json    | 14 +++++---------
 model_prices_and_context_window.json               | 14 +++++---------
 tests/test_litellm/test_utils.py                   | 11 +++++++----
 3 files changed, 17 insertions(+), 22 deletions(-)

diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json
index 0bc6c59a4b7..5bdf6ed4fb8 100644
--- a/litellm/model_prices_and_context_window_backup.json
+++ b/litellm/model_prices_and_context_window_backup.json
@@ -45513,7 +45513,7 @@
     },
     "vertex_ai/lyria-002": {
         "litellm_provider": "vertex_ai",
-                    "mode": "audio_speech",
+        "mode": "audio_speech",
         "output_cost_per_image": 0.06,
         "source": "https://cloud.google.com/gemini-enterprise-agent-platform/generative-ai/pricing#lyria",
         "supported_audio_formats": [
@@ -45549,8 +45549,7 @@
             "/v1/audio/speech"
         ],
         "supported_modalities": [
-            "text",
-            "image"
+            "text"
         ],
         "supported_output_modalities": [
             "audio"
@@ -45561,11 +45560,10 @@
         "supports_audio_input": false,
         "supports_audio_output": true,
         "supports_function_calling": false,
-        "supports_image_input": true,
         "supports_prompt_caching": false,
         "supports_response_schema": false,
         "supports_system_messages": false,
-        "supports_vision": true,
+        "supports_vision": false,
         "supports_web_search": false,
         "vertex_ai_audio_api": "lyria_interactions"
     },
@@ -45588,8 +45586,7 @@
             "/v1/audio/speech"
         ],
         "supported_modalities": [
-            "text",
-            "image"
+            "text"
         ],
         "supported_output_modalities": [
             "audio"
@@ -45600,11 +45597,10 @@
         "supports_audio_input": false,
         "supports_audio_output": true,
         "supports_function_calling": false,
-        "supports_image_input": true,
         "supports_prompt_caching": false,
         "supports_response_schema": false,
         "supports_system_messages": false,
-        "supports_vision": true,
+        "supports_vision": false,
         "supports_web_search": false,
         "vertex_ai_audio_api": "lyria_interactions"
     },
diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json
index 0bc6c59a4b7..5bdf6ed4fb8 100644
--- a/model_prices_and_context_window.json
+++ b/model_prices_and_context_window.json
@@ -45513,7 +45513,7 @@
     },
     "vertex_ai/lyria-002": {
         "litellm_provider": "vertex_ai",
-                    "mode": "audio_speech",
+        "mode": "audio_speech",
         "output_cost_per_image": 0.06,
         "source": "https://cloud.google.com/gemini-enterprise-agent-platform/generative-ai/pricing#lyria",
         "supported_audio_formats": [
@@ -45549,8 +45549,7 @@
             "/v1/audio/speech"
         ],
         "supported_modalities": [
-            "text",
-            "image"
+            "text"
         ],
         "supported_output_modalities": [
             "audio"
@@ -45561,11 +45560,10 @@
         "supports_audio_input": false,
         "supports_audio_output": true,
         "supports_function_calling": false,
-        "supports_image_input": true,
         "supports_prompt_caching": false,
         "supports_response_schema": false,
         "supports_system_messages": false,
-        "supports_vision": true,
+        "supports_vision": false,
         "supports_web_search": false,
         "vertex_ai_audio_api": "lyria_interactions"
     },
@@ -45588,8 +45586,7 @@
             "/v1/audio/speech"
         ],
         "supported_modalities": [
-            "text",
-            "image"
+            "text"
         ],
         "supported_output_modalities": [
             "audio"
@@ -45600,11 +45597,10 @@
         "supports_audio_input": false,
         "supports_audio_output": true,
         "supports_function_calling": false,
-        "supports_image_input": true,
         "supports_prompt_caching": false,
         "supports_response_schema": false,
         "supports_system_messages": false,
-        "supports_vision": true,
+        "supports_vision": false,
         "supports_web_search": false,
         "vertex_ai_audio_api": "lyria_interactions"
     },
diff --git a/tests/test_litellm/test_utils.py b/tests/test_litellm/test_utils.py
index f934b2aca1e..d0916adce9d 100644
--- a/tests/test_litellm/test_utils.py
+++ b/tests/test_litellm/test_utils.py
@@ -2887,14 +2887,17 @@ def test_vertex_ai_lyria_models_in_cost_map():
         "/v1beta/interactions",
         "/v1/audio/speech",
     ]
-    assert clip["supported_modalities"] == ["text", "image"]
-    assert pro["supported_modalities"] == ["text", "image"]
+    assert clip["supported_modalities"] == ["text"]
+    assert pro["supported_modalities"] == ["text"]
+    assert clip["supports_vision"] is False
+    assert pro["supports_vision"] is False
+    assert "supports_image_input" not in clip
+    assert "supports_image_input" not in pro
     assert clip["supported_regions"] == ["global"]
     assert pro["supported_regions"] == ["global"]
     assert clip["supports_audio_output"] is True
     assert pro["supports_audio_output"] is True
-    assert clip["supports_image_input"] is True
-    assert pro["supports_image_input"] is True
+
 
 def test_model_info_for_fireworks_short_form_models():
     """