BerriAI/litellm · warning · ValueError

Cannot specify both maxChunkCount and maxDocumentCount.

Error message

Cannot specify both maxChunkCount and maxDocumentCount.

What it means

GroundingSearchConfig, part of the SAP grounding module config, allows limiting search by maxChunkCount or maxDocumentCount but not both - they are mutually exclusive per the SAP Generative AI Hub API. A Pydantic model_validator (mode='after') raises this ValueError when both fields are non-None.

Source

Thrown at litellm/llms/sap/chat/models.py:153

class KeyValueListPair(BaseModel):
    key: str
    value: list[str]


class DocumentMetadataKeyValueListPairs(KeyValueListPair):
    select_mode: list[Literal["ignoreIfKeyAbsent"]] | None = None


class GroundingSearchConfig(BaseModel):
    max_chunk_count: int | None = Field(default=None, ge=0)
    max_document_count: int | None = Field(default=None, ge=0)

    @model_validator(mode="after")
    def validate_max_chunk_count_and_max_document_count(self):
        if self.max_chunk_count is not None and self.max_document_count is not None:
            raise ValueError("Cannot specify both maxChunkCount and maxDocumentCount.")
        return self


class DocumentGroundingFilter(BaseModel):
    id_: str | None = Field(default=None, alias="id")
    data_repository_type: Literal["vector", "help.sap.com"]
    search_config: GroundingSearchConfig | None = None
    data_repositories: list[str] | None = None
    data_repository_metadata: list[KeyValueListPair] | None = None
    document_metadata: list[DocumentMetadataKeyValueListPairs] | None = None
    chunk_metadata: list[KeyValueListPair] | None = None


class DocumentGroundingPlaceholders(BaseModel):
    input: list[str] = Field(min_length=1)
    output: str

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Keep exactly one of the two limits; delete the other from the config dict.
2. If you need tighter cost control, prefer max_document_count for broad recall budgets or max_chunk_count for per-chunk granularity.
3. Validate configs with the Pydantic model in a unit test before deploying request templates.

Example fix

# before
GroundingSearchConfig(max_chunk_count=5, max_document_count=10)
# after
GroundingSearchConfig(max_document_count=10)

def build_search_config(max_chunk_count=None, max_document_count=None) -> dict:
    if max_chunk_count is not None and max_document_count is not None:
        raise ValueError('set only one of max_chunk_count / max_document_count')
    cfg = {}
    if max_chunk_count is not None:
        cfg['max_chunk_count'] = max_chunk_count
    if max_document_count is not None:
        cfg['max_document_count'] = max_document_count
    return cfg

When it happens

Trigger: Building a DocumentGroundingFilter with GroundingSearchConfig(max_chunk_count=N, max_document_count=M) - both set - in the grounding module config of a sap/ completion request.

Common situations: Copy-pasting full search-config examples from SAP docs into code; teams tuning grounding costs and setting both limits 'to be safe'; config generators that emit every optional field.

Related errors

• For SAP Masking Module Config you must set exactly one of: '
• Content must be a string
• For SAP Masking Module Config you must provide 'providers'.
• For using SAP Filtering Module you must provide at least one
• TranslationModuleConfig requires at least one of 'input' or

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/84d245668eafbb5d. Report an issue: GitHub.