Merge a686f36f0c into 74050e03c5

litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py

@@ -2,6 +2,7 @@
 ## httpx client for vertex ai calls
 ## Initial implementation - covers gemini + image gen calls
 import json
+import re
 import time
 from collections.abc import Callable, Mapping, Sequence
 from copy import deepcopy
@@ -108,6 +109,17 @@ else:
     StreamingChoices = Any
 
 
+GEMINI_MAJOR_VERSION_PATTERN: Final[re.Pattern[str]] = re.compile(r"gemini-(\d+)")
+
+GEMINI_ROLLING_LATEST_ALIASES: Final[frozenset[str]] = frozenset(
+    {
+        "gemini-flash-latest",
+        "gemini-flash-lite-latest",
+        "gemini-pro-latest",
+    }
+)
+
+
 class VertexAIBaseConfig:
     def get_mapped_special_auth_params(self) -> dict:
         """
@@ -259,20 +271,18 @@ class VertexGeminiConfig(VertexAIBaseConfig, BaseConfig):
     @staticmethod
     def _is_gemini_3_or_newer(model: str) -> bool:
         """
-        Check if the model is Gemini 3 Pro or newer.
+        Check if the model is Gemini 3 or newer, e.g. gemini-3-pro-preview,
+        gemini-3.1-flash, gemini-3.5-flash, and every later major version.
 
-        Gemini 3 models include:
-        - gemini-3-pro-preview
-        - gemini-3-flash
-        - gemini-3-flash-preview (Gemini 3 Flash)
-        - gemini-3.1-pro-preview, gemini-3.1-flash, gemini-3.1-flash-lite-preview
-        - gemini-3.5-flash
-        - Any future Gemini 3.x models
+        The major version is read off the model name, so Gemini 4 and beyond
+        satisfy this without a code change. Names carrying no version at all are
+        matched against GEMINI_ROLLING_LATEST_ALIASES, which always resolve to
+        the newest release of their tier.
         """
-        # Check for Gemini 3 models
-        if "gemini-3" in model:
-            return True
-        return False
+        major_version: Final = GEMINI_MAJOR_VERSION_PATTERN.search(model)
+        if major_version is not None:
+            return int(major_version.group(1)) >= 3
+        return model.split("/")[-1] in GEMINI_ROLLING_LATEST_ALIASES
 
     @staticmethod
     def _forward_gemini_function_call_id(model: str) -> bool:

tests/test_litellm/llms/vertex_ai/gemini/test_vertex_and_google_ai_studio_gemini.py

@@ -2431,6 +2431,79 @@ def test_is_gemini_3_or_newer():
     # Edge cases
     assert VertexGeminiConfig._is_gemini_3_or_newer("") == False
 
+    assert VertexGeminiConfig._is_gemini_3_or_newer("gemini-flash-latest") == True
+    assert VertexGeminiConfig._is_gemini_3_or_newer("gemini-flash-lite-latest") == True
+    assert VertexGeminiConfig._is_gemini_3_or_newer("gemini-pro-latest") == True
+    assert (
+        VertexGeminiConfig._is_gemini_3_or_newer("gemini/gemini-flash-latest") == True
+    )
+    assert (
+        VertexGeminiConfig._is_gemini_3_or_newer("vertex_ai/gemini-pro-latest") == True
+    )
+
+    assert (
+        VertexGeminiConfig._is_gemini_3_or_newer(
+            "gemini-2.5-flash-native-audio-latest"
+        )
+        == False
+    )
+
+    # Later major versions must satisfy this without a code change
+    assert VertexGeminiConfig._is_gemini_3_or_newer("gemini-4-flash") == True
+    assert VertexGeminiConfig._is_gemini_3_or_newer("gemini-4.2-pro-preview") == True
+    assert VertexGeminiConfig._is_gemini_3_or_newer("gemini/gemini-10-flash") == True
+
+    # Unversioned names that are not rolling aliases stay excluded
+    assert VertexGeminiConfig._is_gemini_3_or_newer("gemini-embedding-001") == False
+    assert VertexGeminiConfig._is_gemini_3_or_newer("gemini-exp-1206") == False
+
+
+def test_thought_signature_fallback_for_rolling_latest_alias():
+    """
+    A tool call whose thought signature did not survive the round-trip must
+    still get the dummy signature Google documents, otherwise Gemini rejects
+    the follow-up turn with:
+
+      400 Function call is missing a thought_signature in functionCall parts.
+
+    Regression test: `gemini-flash-latest` serves Gemini 3.x but was not
+    detected as such, so the fallback was skipped and every multi-turn tool
+    call that lost its signature failed.
+    """
+    from litellm.litellm_core_utils.prompt_templates.factory import (
+        convert_to_gemini_tool_call_invoke,
+    )
+
+    message = {
+        "role": "assistant",
+        "content": None,
+        "tool_calls": [
+            {
+                "id": "call_abc123",  # no `__thought__` suffix -> signature lost
+                "type": "function",
+                "function": {
+                    "name": "search_website",
+                    "arguments": '{"query": "Iceland"}',
+                },
+            }
+        ],
+    }
+
+    for model in [
+        "gemini-3-flash",
+        "gemini-flash-latest",
+        "gemini-flash-lite-latest",
+        "gemini-pro-latest",
+        "gemini/gemini-flash-latest",
+    ]:
+        parts = convert_to_gemini_tool_call_invoke(message, model=model)
+        assert any(
+            "thoughtSignature" in part for part in parts
+        ), f"expected a thought signature for {model}"
+
+    parts = convert_to_gemini_tool_call_invoke(message, model="gemini-2.5-flash")
+    assert not any("thoughtSignature" in part for part in parts)
+
 
 def _tool_call_messages(tool_call_id: str):
     return [