Merge pull request #4313 from BerriAI/litellm_drop_specific_params

fix(utils.py): allow dropping specific openai params

litellm/proxy/utils.py

@@ -691,7 +691,7 @@ class PrismaClient:
             finally:
                 os.chdir(original_dir)
             # Now you can import the Prisma Client
-            from prisma import Prisma
+            from prisma import Prisma  # type: ignore
 
         self.db = Prisma()  # Client to connect to Prisma db
 

litellm/tests/test_optional_params.py

@@ -1,20 +1,25 @@
 #### What this tests ####
 #    This tests if get_optional_params works as expected
-import sys, os, time, inspect, asyncio, traceback
+import asyncio
+import inspect
+import os
+import sys
+import time
+import traceback
+
 import pytest
 
 sys.path.insert(0, os.path.abspath("../.."))
+from unittest.mock import MagicMock, patch
+
 import litellm
-from litellm.utils import get_optional_params_embeddings, get_optional_params
+from litellm.llms.prompt_templates.factory import map_system_message_pt
-from litellm.llms.prompt_templates.factory import (
-    map_system_message_pt,
-)
-from unittest.mock import patch, MagicMock
 from litellm.types.completion import (
-    ChatCompletionUserMessageParam,
-    ChatCompletionSystemMessageParam,
     ChatCompletionMessageParam,
+    ChatCompletionSystemMessageParam,
+    ChatCompletionUserMessageParam,
 )
+from litellm.utils import get_optional_params, get_optional_params_embeddings
 
 ## get_optional_params_embeddings
 ### Models: OpenAI, Azure, Bedrock
@@ -286,3 +291,45 @@ def test_dynamic_drop_params_e2e():
         mock_response.assert_called_once()
         print(mock_response.call_args.kwargs["data"])
         assert "response_format" not in mock_response.call_args.kwargs["data"]
+
+
+@pytest.mark.parametrize("drop_params", [True, False, None])
+def test_dynamic_drop_additional_params(drop_params):
+    """
+    Make a call to cohere, dropping 'response_format' specifically
+    """
+    if drop_params is True:
+        optional_params = litellm.utils.get_optional_params(
+            model="command-r",
+            custom_llm_provider="cohere",
+            response_format="json",
+            additional_drop_params=["response_format"],
+        )
+    else:
+        try:
+            optional_params = litellm.utils.get_optional_params(
+                model="command-r",
+                custom_llm_provider="cohere",
+                response_format="json",
+            )
+            pytest.fail("Expected to fail")
+        except Exception as e:
+            pass
+
+
+def test_dynamic_drop_additional_params_e2e():
+    with patch("requests.post", new=MagicMock()) as mock_response:
+        try:
+            response = litellm.completion(
+                model="command-r",
+                messages=[{"role": "user", "content": "Hey, how's it going?"}],
+                response_format={"key": "value"},
+                additional_drop_params=["response_format"],
+            )
+        except Exception as e:
+            pass
+
+        mock_response.assert_called_once()
+        print(mock_response.call_args.kwargs["data"])
+        assert "response_format" not in mock_response.call_args.kwargs["data"]
+        assert "additional_drop_params" not in mock_response.call_args.kwargs["data"]

litellm/utils.py

@@ -2293,6 +2293,7 @@ def get_optional_params(
     extra_headers=None,
     api_version=None,
     drop_params=None,
+    additional_drop_params=None,
     **kwargs,
 ):
     # retrieve all parameters passed to the function
@@ -2309,7 +2310,6 @@ def get_optional_params(
             k.startswith("vertex_") and custom_llm_provider != "vertex_ai"
         ):  # allow dynamically setting vertex ai init logic
             continue
-
         passed_params[k] = v
 
     optional_params: Dict = {}
@@ -2365,7 +2365,19 @@ def get_optional_params(
         "extra_headers": None,
         "api_version": None,
         "drop_params": None,
+        "additional_drop_params": None,
     }
+
+    def _should_drop_param(k, additional_drop_params) -> bool:
+        if (
+            additional_drop_params is not None
+            and isinstance(additional_drop_params, list)
+            and k in additional_drop_params
+        ):
+            return True  # allow user to drop specific params for a model - e.g. vllm - logit bias
+
+        return False
+
     # filter out those parameters that were passed with non-default values
     non_default_params = {
         k: v
@@ -2375,8 +2387,11 @@ def get_optional_params(
             and k != "custom_llm_provider"
             and k != "api_version"
             and k != "drop_params"
+            and k != "additional_drop_params"
             and k in default_params
             and v != default_params[k]
+            and _should_drop_param(k=k, additional_drop_params=additional_drop_params)
+            is False
         )
     }