[Feat] New model vertex_ai/deepseek-ai/deepseek-r1-0528-maas (#13594)

* add ertex_ai/deepseek-ai/deepseek-r1-0528-maas

* fix init

* test_model_info_for_vertex_ai_deepseek_model

litellm/__init__.py

@@ -486,6 +486,7 @@ vertex_code_text_models: List = []
 vertex_embedding_models: List = []
 vertex_anthropic_models: List = []
 vertex_llama3_models: List = []
+vertex_deepseek_models: List = []
 vertex_ai_ai21_models: List = []
 vertex_mistral_models: List = []
 ai21_models: List = []
@@ -618,6 +619,9 @@ def add_known_models():
         elif value.get("litellm_provider") == "vertex_ai-llama_models":
             key = key.replace("vertex_ai/", "")
             vertex_llama3_models.append(key)
+        elif value.get("litellm_provider") == "vertex_ai-deepseek_models":
+            key = key.replace("vertex_ai/", "")
+            vertex_deepseek_models.append(key)
         elif value.get("litellm_provider") == "vertex_ai-mistral_models":
             key = key.replace("vertex_ai/", "")
             vertex_mistral_models.append(key)
@@ -850,7 +854,8 @@ models_by_provider: dict = {
     + vertex_text_models
     + vertex_anthropic_models
     + vertex_vision_models
-    + vertex_language_models,
+    + vertex_language_models
+    + vertex_deepseek_models,
     "ai21": ai21_models,
     "bedrock": bedrock_models + bedrock_converse_models,
     "petals": petals_models,

litellm/model_prices_and_context_window_backup.json

@@ -9686,6 +9686,21 @@
         "supports_assistant_prefill": true,
         "supports_tool_choice": true
     },
+    "vertex_ai/deepseek-ai/deepseek-r1-0528-maas": {
+        "max_tokens": 8192,
+        "max_input_tokens": 65336,
+        "max_output_tokens": 8192,
+        "input_cost_per_token": 1.35e-06,
+        "output_cost_per_token": 5.4e-06,
+        "litellm_provider": "vertex_ai-deepseek_models",
+        "mode": "chat",
+        "source": "https://cloud.google.com/vertex-ai/generative-ai/pricing#partner-models",
+        "supports_function_calling": true,
+        "supports_assistant_prefill": true,
+        "supports_reasoning": true,
+        "supports_tool_choice": true,
+        "supports_prompt_caching": true
+    },
     "vertex_ai/meta/llama3-405b-instruct-maas": {
         "max_tokens": 32000,
         "max_input_tokens": 32000,

model_prices_and_context_window.json

@@ -9686,6 +9686,21 @@
         "supports_assistant_prefill": true,
         "supports_tool_choice": true
     },
+    "vertex_ai/deepseek-ai/deepseek-r1-0528-maas": {
+        "max_tokens": 8192,
+        "max_input_tokens": 65336,
+        "max_output_tokens": 8192,
+        "input_cost_per_token": 1.35e-06,
+        "output_cost_per_token": 5.4e-06,
+        "litellm_provider": "vertex_ai-deepseek_models",
+        "mode": "chat",
+        "source": "https://cloud.google.com/vertex-ai/generative-ai/pricing#partner-models",
+        "supports_function_calling": true,
+        "supports_assistant_prefill": true,
+        "supports_reasoning": true,
+        "supports_tool_choice": true,
+        "supports_prompt_caching": true
+    },
     "vertex_ai/meta/llama3-405b-instruct-maas": {
         "max_tokens": 32000,
         "max_input_tokens": 32000,

tests/test_litellm/test_utils.py

@@ -2416,3 +2416,16 @@ def test_generate_gcp_iam_access_token_import_error():
 if __name__ == "__main__":
     # Allow running this test file directly for debugging
     pytest.main([__file__, "-v"])
+
+
+def test_model_info_for_vertex_ai_deepseek_model():
+    model_info = litellm.get_model_info(
+        model="vertex_ai/deepseek-ai/deepseek-r1-0528-maas"
+    )
+    assert model_info is not None
+    assert model_info["litellm_provider"] == "vertex_ai-deepseek_models"
+    assert model_info["mode"] == "chat"
+
+    assert model_info["input_cost_per_token"] is not None
+    assert model_info["output_cost_per_token"] is not None
+    print("vertex deepseek model info", model_info)