scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-21 00:21:49 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 74

fix(gemini): preserve candidates with finishReason and no content (#40477)

Browse source
This commit is contained in:
tusharjamunkar 2026-09-12 22:16:19 +05:30
parent 1c61c2606e
commit 97dbd2dfbf
5 changed files with 184 additions and 15 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
litellm/litellm_core_utils/core_helpers.py
View file

@ -224,6 +224,7 @@ _FINISH_REASON_MAP: Final[dict[str, OpenAIChatCompletionFinishReason]] = {
"IMAGE_PROHIBITED_CONTENT": "content_filter",
"TOO_MANY_TOOL_CALLS": "stop",
"MALFORMED_RESPONSE": "stop",
"NO_IMAGE": "content_filter",
# Zhipu GLM
"network_error": "stop",
"sensitive": "content_filter",

2
litellm/llms/anthropic/experimental_pass_through/adapters/transformation.py
View file

@ -1367,6 +1367,8 @@ class LiteLLMAnthropicMessagesAdapter:
return "max_tokens"
elif openai_finish_reason == "tool_calls":
return "tool_use"
elif openai_finish_reason in ["content_filter", "refusal"]:
return "refusal"
return "end_turn"
@staticmethod

30
litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py
View file

@ -1340,6 +1340,7 @@ class VertexGeminiConfig(VertexAIBaseConfig, BaseConfig):
"IMAGE_PROHIBITED_CONTENT",
"TOO_MANY_TOOL_CALLS",
"MALFORMED_RESPONSE",
"NO_IMAGE",
}
)
@ -2224,22 +2225,23 @@ class VertexGeminiConfig(VertexAIBaseConfig, BaseConfig):
grounding_metadata: Final[list[dict]] = []
url_context_metadata: Final[list[dict]] = []
image_response: list[ImageURLListItem] | None = None
safety_ratings: Final[list] = []
citation_metadata: Final[list] = []
chat_completion_message: Final[ChatCompletionResponseMessage] = {"role": "assistant"}
chat_completion_logprobs: ChoiceLogprobs | None = None
tools: list[ChatCompletionToolCallChunk] | None = []
functions: ChatCompletionToolCallFunctionChunk | None = None
thinking_blocks: list[ChatCompletionThinkingBlock] | None = None
reasoning_content: str | None = None
thought_signatures: Sequence[str] | None = None
server_side_tool_invocations: list[dict[str, object]] | None = None
for idx, candidate in enumerate(_candidates):
if "content" not in candidate:
if "content" not in candidate and "finishReason" not in candidate:
continue
image_response: list[ImageURLListItem] | None = None
chat_completion_message: ChatCompletionResponseMessage = {"role": "assistant"}
chat_completion_logprobs: ChoiceLogprobs | None = None
tools: list[ChatCompletionToolCallChunk] | None = []
functions: ChatCompletionToolCallFunctionChunk | None = None
thinking_blocks: list[ChatCompletionThinkingBlock] | None = None
reasoning_content: str | None = None
thought_signatures: Sequence[str] | None = None
server_side_tool_invocations: list[dict[str, object]] | None = None
# Extract metadata using helper function
(
candidate_grounding_metadata,
@ -2253,7 +2255,7 @@ class VertexGeminiConfig(VertexAIBaseConfig, BaseConfig):
safety_ratings.extend(candidate_safety_ratings)
citation_metadata.extend(candidate_citation_metadata)
if "parts" in candidate["content"]:
if "content" in candidate and candidate["content"] and "parts" in candidate["content"]:
(
content,
reasoning_content,
@ -2348,6 +2350,11 @@ class VertexGeminiConfig(VertexAIBaseConfig, BaseConfig):
tool_invocation_fields["server_side_tool_invocations"] = server_side_tool_invocations
chat_completion_message["provider_specific_fields"] = tool_invocation_fields
if candidate.get("finishReason"):
finish_reason_fields = chat_completion_message.get("provider_specific_fields") or {}
finish_reason_fields["native_finish_reason"] = candidate.get("finishReason")
chat_completion_message["provider_specific_fields"] = finish_reason_fields
if isinstance(model_response, ModelResponseStream):
choice = VertexGeminiConfig._create_streaming_choice(
chat_completion_message=chat_completion_message,
@ -2368,6 +2375,7 @@ class VertexGeminiConfig(VertexAIBaseConfig, BaseConfig):
message=chat_completion_message,
logprobs=chat_completion_logprobs,
enhancements=None,
provider_specific_fields=chat_completion_message.get("provider_specific_fields"),
)
model_response.choices.append(choice)

20
litellm/responses/litellm_completion_transformation/transformation.py
View file

@ -2272,13 +2272,27 @@ class LiteLLMCompletionResponsesConfig:
if choices and len(choices) > 0:
finish_reason = choices[0].finish_reason
status: Final[ResponsesAPIStatus] = (
LiteLLMCompletionResponsesConfig._map_chat_completion_finish_reason_to_responses_status(
finish_reason
)
)
incomplete_details = getattr(chat_completion_response, "incomplete_details", None)
if incomplete_details is None and status == "incomplete":
from openai.types.responses.response import IncompleteDetails
if finish_reason == "length":
incomplete_details = IncompleteDetails(reason="max_output_tokens")
elif finish_reason in ["content_filter", "refusal"]:
incomplete_details = IncompleteDetails(reason="content_filter")
responses_api_response: Final[ResponsesAPIResponse] = ResponsesAPIResponse(
id=chat_completion_response.id,
created_at=chat_completion_response.created,
model=chat_completion_response.model,
object="response",
error=getattr(chat_completion_response, "error", None),
incomplete_details=getattr(chat_completion_response, "incomplete_details", None),
incomplete_details=incomplete_details,
instructions=getattr(chat_completion_response, "instructions", None),
metadata=getattr(chat_completion_response, "metadata", {}),
output=LiteLLMCompletionResponsesConfig._transform_chat_completion_choices_to_responses_output(
@ -2296,9 +2310,7 @@ class LiteLLMCompletionResponsesConfig:
max_output_tokens=getattr(chat_completion_response, "max_output_tokens", None),
previous_response_id=getattr(chat_completion_response, "previous_response_id", None),
reasoning=None,
status=LiteLLMCompletionResponsesConfig._map_chat_completion_finish_reason_to_responses_status(
finish_reason
),
status=status,
text={},
truncation=getattr(chat_completion_response, "truncation", None),
usage=LiteLLMCompletionResponsesConfig._transform_chat_completion_usage_to_responses_usage(

146
tests/test_litellm/llms/vertex_ai/gemini/test_vertex_and_google_ai_studio_gemini.py
View file

@ -5836,3 +5836,149 @@ def test_supported_reasoning_efforts_still_map(model):
drop_params=False,
)
assert "thinkingConfig" in result
def test_gemini_candidate_with_finish_reason_no_content_chat_completion():
config = VertexGeminiConfig()
completion_response = {
"candidates": [
{
"finishReason": "NO_IMAGE",
"index": 0,
}
],
"usageMetadata": {
"promptTokenCount": 19,
"candidatesTokenCount": 0,
"totalTokenCount": 19,
},
}
model_response = ModelResponse()
logging_obj = MagicMock()
raw_response = MagicMock()
raw_response.headers = {}
resp = config._transform_google_generate_content_to_openai_model_response(
completion_response=completion_response,
model_response=model_response,
model="gemini-2.5-flash-image",
logging_obj=logging_obj,
raw_response=raw_response,
)
assert len(resp.choices) == 1
assert resp.choices[0].finish_reason == "content_filter"
assert resp.choices[0].message.content is None
assert resp.choices[0].provider_specific_fields["native_finish_reason"] == "NO_IMAGE"
def test_gemini_candidate_with_finish_reason_no_content_anthropic_messages():
from litellm.llms.anthropic.experimental_pass_through.adapters.transformation import (
LiteLLMAnthropicMessagesAdapter,
)
config = VertexGeminiConfig()
completion_response = {
"candidates": [
{
"finishReason": "NO_IMAGE",
"index": 0,
}
],
"usageMetadata": {
"promptTokenCount": 19,
"candidatesTokenCount": 0,
"totalTokenCount": 19,
},
}
resp = config._transform_google_generate_content_to_openai_model_response(
completion_response=completion_response,
model_response=ModelResponse(),
model="gemini-2.5-flash-image",
logging_obj=MagicMock(),
raw_response=MagicMock(headers={}),
)
adapter = LiteLLMAnthropicMessagesAdapter()
anthropic_resp = adapter.translate_openai_response_to_anthropic(
response=resp,
tool_name_mapping={},
)
assert anthropic_resp["stop_reason"] == "refusal"
assert anthropic_resp["content"] == []
def test_gemini_candidate_with_finish_reason_no_content_responses_api():
from litellm.responses.litellm_completion_transformation.transformation import (
LiteLLMCompletionResponsesConfig,
)
config = VertexGeminiConfig()
completion_response = {
"candidates": [
{
"finishReason": "NO_IMAGE",
"index": 0,
}
],
"usageMetadata": {
"promptTokenCount": 19,
"candidatesTokenCount": 0,
"totalTokenCount": 19,
},
}
resp = config._transform_google_generate_content_to_openai_model_response(
completion_response=completion_response,
model_response=ModelResponse(),
model="gemini-2.5-flash-image",
logging_obj=MagicMock(),
raw_response=MagicMock(headers={}),
)
responses_resp = LiteLLMCompletionResponsesConfig.transform_chat_completion_response_to_responses_api_response(
request_input="Generate picture",
responses_api_request={},
chat_completion_response=resp,
)
assert responses_resp.status == "incomplete"
assert responses_resp.incomplete_details is not None
assert responses_resp.incomplete_details.reason == "content_filter"
def test_gemini_candidate_other_finish_reasons_no_content():
from litellm.llms.anthropic.experimental_pass_through.adapters.transformation import (
LiteLLMAnthropicMessagesAdapter,
)
from litellm.responses.litellm_completion_transformation.transformation import (
LiteLLMCompletionResponsesConfig,
)
config = VertexGeminiConfig()
max_tokens_response = {
"candidates": [{"finishReason": "MAX_TOKENS", "index": 0}],
"usageMetadata": {"promptTokenCount": 10, "candidatesTokenCount": 50, "totalTokenCount": 60},
}
resp_length = config._transform_google_generate_content_to_openai_model_response(
completion_response=max_tokens_response,
model_response=ModelResponse(),
model="gemini-2.5-flash",
logging_obj=MagicMock(),
raw_response=MagicMock(headers={}),
)
assert len(resp_length.choices) == 1
assert resp_length.choices[0].finish_reason == "length"
assert resp_length.choices[0].provider_specific_fields["native_finish_reason"] == "MAX_TOKENS"
anthropic_length = LiteLLMAnthropicMessagesAdapter().translate_openai_response_to_anthropic(
response=resp_length,
tool_name_mapping={},
)
assert anthropic_length["stop_reason"] == "max_tokens"
responses_length = LiteLLMCompletionResponsesConfig.transform_chat_completion_response_to_responses_api_response(
request_input="thinking request",
responses_api_request={},
chat_completion_response=resp_length,
)
assert responses_length.status == "incomplete"
assert responses_length.incomplete_details.reason == "max_output_tokens"