invoke-ai/InvokeAI · error · ValueError

Either bounding_box or points must be provided

Error message

Either bounding_box or points must be provided

What it means

SAMInput is a pydantic model that must contain at least one segmentation prompt: a bounding_box or a list of points. The model_validator `check_input` raises this error when both are None/empty. SAM cannot run without at least one prompt indicating where to segment.

Source

Thrown at invokeai/backend/image_util/segment_anything/shared.py:48

    negative = -1
    neutral = 0
    positive = 1


class SAMPoint(BaseModel):
    x: int = Field(..., description="The x-coordinate of the point")
    y: int = Field(..., description="The y-coordinate of the point")
    label: SAMPointLabel = Field(..., description="The label of the point")


class SAMInput(BaseModel):
    bounding_box: BoundingBox | None = Field(None, description="The bounding box to use for segmentation")
    points: list[SAMPoint] | None = Field(None, description="The points to use for segmentation")

    @model_validator(mode="after")
    def check_input(self):
        if not self.bounding_box and not self.points:
            raise ValueError("Either bounding_box or points must be provided")
        return self

View on GitHub (pinned to 0b6a024f2f)

Solutions

1. Provide a bounding_box, e.g. SAMInput(bounding_box=BoundingBox(x_min=..., x_max=..., y_min=..., y_max=...)).
2. Or provide at least one SAMPoint, e.g. SAMInput(points=[SAMPoint(x=..., y=..., label=SAMPointLabel.positive)]).
3. If the prompt comes from user input, validate that a selection exists before constructing SAMInput and surface a user-facing message instead.

Example fix

// before
SAMInput()  # or SAMInput(bounding_box=None, points=[])
// after
SAMInput(points=[SAMPoint(x=250, y=180, label=SAMPointLabel.positive)])

def validate_sam_input(payload: dict) -> dict | None:
    if not payload.get("bounding_box") and not payload.get("points"):
        return None
    return payload

def has_prompt(s: SAMInput) -> bool:
    return bool(s.bounding_box) or bool(s.points)

try:
    sam_input = SAMInput(**payload)
except ValueError as e:
    raise UserInputError("Please select a region or at least one point before running segmentation.") from e

When it happens

Trigger: Constructing SAMInput() with neither field set, passing points=[] (empty list is falsy), passing bounding_box=None and omitting points, or deserializing JSON that omits both keys.

Common situations: Building the request programmatically where a UI selection was empty, conditionally assembling a payload and dropping both prompt fields, or an upstream API returning null for the region of interest.

Related errors

• stop must be greater than start
• cfg_scale must be greater than 1
• Face IDs must be a comma-separated list of integers (e.g. "1
• cfg_scale values must be finite.
• shift must be finite.

AI-assisted analysis of invoke-ai/InvokeAI@0b6a024f2f (2026-08-29). Data as JSON: /api/errors/27190e18493e540e. Report an issue: GitHub.