run-llama/llama_index · error · ValueError

questions must be >= 1

Error message

questions must be >= 1

What it means

Raised by QuestionsAnsweredExtractor.__init__ when the questions parameter (number of questions the LLM should generate per node, default 5) is less than 1. Like the other metadata extractors, the count is validated eagerly in the constructor because it is interpolated into prompt_template.

Source

Thrown at llama-index-core/llama_index/core/extractors/metadata_extractors.py:308

    )
    embedding_only: bool = Field(
        default=True, description="Whether to use metadata for emebddings only."
    )

    def __init__(
        self,
        llm: Optional[LLM] = None,
        # TODO: llm_predictor arg is deprecated
        llm_predictor: Optional[LLM] = None,
        questions: int = 5,
        prompt_template: str = DEFAULT_QUESTION_GEN_TMPL,
        embedding_only: bool = True,
        num_workers: int = DEFAULT_NUM_WORKERS,
        **kwargs: Any,
    ) -> None:
        """Init params."""
        if questions < 1:
            raise ValueError("questions must be >= 1")

        super().__init__(
            llm=llm or llm_predictor or Settings.llm,
            questions=questions,
            prompt_template=prompt_template,
            embedding_only=embedding_only,
            num_workers=num_workers,
            **kwargs,
        )

    @classmethod
    def class_name(cls) -> str:
        return "QuestionsAnsweredExtractor"

    async def _aextract_questions_from_node(self, node: BaseNode) -> Dict[str, str]:
        """Extract questions from a node and return it's metadata dict."""
        if self.is_text_node_only and not isinstance(node, TextNode):
            return {}

View on GitHub (pinned to afd0fef371)

Solutions

1. Pass questions >= 1, e.g. QuestionsAnsweredExtractor() for the default of 5.
2. Clamp config inputs: questions = max(1, cfg['questions']).
3. If question metadata is unwanted, drop the extractor from the pipeline.

Example fix

# before
extractor = QuestionsAnsweredExtractor(questions=cfg["questions"])  # 0

# after
extractor = QuestionsAnsweredExtractor(questions=max(1, cfg["questions"]))

num_questions = int(config.get("questions", 5))
if num_questions < 1:
    raise ValueError(f"questions must be >= 1, got {num_questions}")
extractor = QuestionsAnsweredExtractor(questions=num_questions)

When it happens

Trigger: Constructing QuestionsAnsweredExtractor(questions=0) from a misconfigured value — config file, environment variable, or computed expression that produces zero/negative.

Common situations: Shared config templates reused across extractors where questions defaulted to 0; dynamic question counts derived from document size; refactors that renamed the old questions_per_chunk parameter and lost the value.

Related errors

• num_nodes must be >= 1
• num_keywords must be >= 1
• All agents must have a name in a multi-agent workflow
• All agents must have a description in a multi-agent workflow
• Initial state is not supported per-agent in AgentWorkflow

AI-assisted analysis of run-llama/llama_index@afd0fef371 (2026-08-15). Data as JSON: /api/errors/c79b30dafd49bd82. Report an issue: GitHub.