vikarti.anatra/Skyvern
1
0
Fork 0
mirror of https://github.com/Skyvern-AI/skyvern.git synced 2026-07-09 16:09:13 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 5

fix(llm): re-point Vertex gemini-3.1-flash-lite at the GA id (preview sunsets 2026-07-09) (#7162)

Browse source 
Co-authored-by: AronPerez <aperez0295@gmail.com>
This commit is contained in:
pedrohsdb 2026-07-07 12:14:46 -07:00 • committed by GitHub
parent ee0964b43a
commit ab94965186
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 6 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

8
skyvern/forge/sdk/api/llm/config_registry.py
View file

@ -1638,13 +1638,13 @@ if settings.ENABLE_VERTEX_AI:
LLMConfigRegistry.register_config(
"VERTEX_GEMINI_3.1_FLASH_LITE",
LLMConfig(
"vertex_ai/gemini-3.1-flash-lite-preview",
"vertex_ai/gemini-3.1-flash-lite",
[],
supports_vision=True,
add_assistant_prefix=False,
max_completion_tokens=65536,
litellm_params=LiteLLMParams(
api_base=f"{api_base}/gemini-3.1-flash-lite-preview" if api_base else None,
api_base=f"{api_base}/gemini-3.1-flash-lite" if api_base else None,
vertex_location=settings.VERTEX_LOCATION,
thinking_level="medium" if settings.GEMINI_INCLUDE_THOUGHT else "minimal",
vertex_credentials=settings.VERTEX_CREDENTIALS,
@ -1757,13 +1757,13 @@ if settings.ENABLE_VERTEX_AI:
LLMConfigRegistry.register_config(
"VERTEX_GEMINI_3.1_FLASH_LITE_FLEX",
LLMConfig(
"vertex_ai/gemini-3.1-flash-lite-preview",
"vertex_ai/gemini-3.1-flash-lite",
[],
supports_vision=True,
add_assistant_prefix=False,
max_completion_tokens=65536,
litellm_params=LiteLLMParams(
api_base=f"{api_base}/gemini-3.1-flash-lite-preview" if api_base else None,
api_base=f"{api_base}/gemini-3.1-flash-lite" if api_base else None,
vertex_location=settings.VERTEX_LOCATION,
thinking_level="medium" if settings.GEMINI_INCLUDE_THOUGHT else "minimal",
vertex_credentials=settings.VERTEX_CREDENTIALS,

4
tests/unit/test_api_handler_factory.py
View file

@ -787,7 +787,7 @@ async def test_router_retries_content_filter_on_first_non_gemini_fallback(monkey
async def acompletion(self, *, model: str, messages: Any, **kwargs: Any) -> FakeLLMResponse:
calls.append(model)
if len(calls) == 1:
return FakeLLMResponse("gemini-3.1-flash-lite-preview", content=None, finish_reason="content_filter")
return FakeLLMResponse("gemini-3.1-flash-lite", content=None, finish_reason="content_filter")
fallback_kwargs.update(kwargs)
return FakeLLMResponse("gpt-5-fallback", content='{"actions": []}')
@ -828,7 +828,7 @@ async def test_router_does_not_retry_content_filter_without_non_gemini_fallback(
async def acompletion(self, *, model: str, messages: Any, **kwargs: Any) -> FakeLLMResponse:
calls.append(model)
return FakeLLMResponse("gemini-3.1-flash-lite-preview", content=None, finish_reason="content_filter")
return FakeLLMResponse("gemini-3.1-flash-lite", content=None, finish_reason="content_filter")
monkeypatch.setattr(api_handler_factory.litellm, "Router", _AlwaysFilterRouter)