fix(llm_cost_calc/google.py): fix google embedding cost calculation

Fixes https://github.com/BerriAI/litellm/issues/4630

litellm/litellm_core_utils/litellm_logging.py

@@ -1275,7 +1275,7 @@ class Logging:
                     f"Model={self.model}; cost={self.model_call_details['response_cost']}"
                 )
             except litellm.NotFoundError as e:
-                verbose_logger.error(
+                verbose_logger.warning(
                     f"Model={self.model} not found in completion cost map. Setting 'response_cost' to None"
                 )
                 self.model_call_details["response_cost"] = None

litellm/litellm_core_utils/llm_cost_calc/google.py

@@ -1,7 +1,7 @@
 # What is this?
 ## Cost calculation for Google AI Studio / Vertex AI models
 import traceback
-from typing import List, Literal, Optional, Tuple
+from typing import List, Literal, Optional, Tuple, Union
 
 import litellm
 from litellm import verbose_logger
@@ -29,6 +29,32 @@ def _is_above_128k(tokens: float) -> bool:
     return False
 
 
+def cost_router(
+    model: str,
+    custom_llm_provider: str,
+    prompt_tokens: float,
+    completion_tokens: float,
+    prompt_characters: float,
+    completion_characters: float,
+    call_type: Union[Literal["embedding", "aembedding"], str],
+) -> Literal["cost_per_character", "cost_per_token"]:
+    """
+    Route the cost calc to the right place, based on model/call_type/etc.
+
+    Returns
+        - str, the specific google cost calc function it should route to.
+    """
+    if custom_llm_provider == "vertex_ai" and "claude" in model:
+        return "cost_per_token"
+    elif custom_llm_provider == "gemini":
+        return "cost_per_token"
+    elif custom_llm_provider == "vertex_ai" and (
+        call_type == "embedding" or call_type == "aembedding"
+    ):
+        return "cost_per_token"
+    return "cost_per_character"
+
+
 def cost_per_character(
     model: str,
     custom_llm_provider: str,