BerriAI/litellm · error · ValueError

Unsupported Amazon Nova Canvas taskType: {task_type!r}. Use

Error message

Unsupported Amazon Nova Canvas taskType: {task_type!r}. Use BACKGROUND_REMOVAL, OUTPAINTING, IMAGE_VARIATION, INPAINTING, or omit taskType for automatic routing (mask → INPAINTING, else IMAGE_VARIATION).

What it means

The Nova Canvas image-edit transformation only accepts taskType values BACKGROUND_REMOVAL, OUTPAINTING, IMAGE_VARIATION, and INPAINTING (or no taskType for automatic routing). Any other non-empty taskType reaching the final validation branch raises this ValueError listing the allowed values.

Source

Thrown at litellm/llms/bedrock/image_edit/amazon_nova_canvas_image_edit_transformation.py:97

    # Honour explicit IMAGE_VARIATION even when a mask is present (mask is ignored
    # for this task type; callers use INPAINTING when they want mask semantics).
    if task_type == "IMAGE_VARIATION":
        var_params_explicit: Final[dict[str, Any]] = {
            "images": [image_b64],
            "text": text,
        }
        if negative_text is not None:
            var_params_explicit["negativeText"] = negative_text
        if similarity_strength is not None:
            var_params_explicit["similarityStrength"] = similarity_strength
        return {
            "taskType": "IMAGE_VARIATION",
            "imageVariationParams": var_params_explicit,
        }
    # Explicit taskType must be INPAINTING or omitted from here on; anything else is invalid.
    if task_type is not None and str(task_type).strip() != "":
        if task_type != "INPAINTING":
            raise ValueError(
                f"Unsupported Amazon Nova Canvas taskType: {task_type!r}. "
                "Use BACKGROUND_REMOVAL, OUTPAINTING, IMAGE_VARIATION, INPAINTING, "
                "or omit taskType for automatic routing (mask → INPAINTING, else IMAGE_VARIATION)."
            )
    if mask_b64 is not None or mask_prompt is not None or task_type == "INPAINTING":
        in_params: Final[dict[str, Any]] = {"image": image_b64, "text": text}
        if mask_prompt is not None:
            in_params["maskPrompt"] = mask_prompt
        elif mask_b64 is not None:
            in_params["maskImage"] = mask_b64
        if negative_text is not None:
            in_params["negativeText"] = negative_text
        if "maskPrompt" not in in_params and "maskImage" not in in_params:
            raise ValueError(
                "Amazon Nova Canvas INPAINTING requires either maskPrompt or maskImage "
                "(use OpenAI mask= for maskImage, or pass maskPrompt in optional params). "
                "See https://docs.aws.amazon.com/nova/latest/userguide/image-gen-req-resp-structure.html"
            )

View on GitHub (pinned to 6c2dcb801b)

Solutions

Set taskType to one of BACKGROUND_REMOVAL, OUTPAINTING, IMAGE_VARIATION, INPAINTING (exact uppercase), or omit it for auto routing.
For background removal, remember only BACKGROUND_REMOVAL needs no mask and no text prompt requirements differ per task.
If you wanted color-guided generation, use the image-generation endpoint (litellm.image / /v1/images/generations), not /v1/images/edits.

Example fix

# before
optional_params={"taskType": "COLOR_GUIDED_GENERATION"}  # edits -> ValueError
# after
optional_params={"taskType": "INPAINTING", "maskPrompt": "the sky"}
# or omit taskType entirely

Defensive patterns

Strategy: validation

Validate before calling

EDIT_TASK_TYPES = {"BACKGROUND_REMOVAL", "OUTPAINTING", "IMAGE_VARIATION", "INPAINTING"}

def valid_edit_task_type(tt: str | None) -> bool:
    return tt is None or (isinstance(tt, str) and tt.strip() in EDIT_TASK_TYPES or tt in EDIT_TASK_TYPES)

Type guard

def is_nova_edit_task_type(tt: object) -> bool:
    return isinstance(tt, str) and tt in {
        "BACKGROUND_REMOVAL", "OUTPAINTING", "IMAGE_VARIATION", "INPAINTING"
    }

Prevention

Keep an enum/constant set of edit task types and validate against it client-side.
Remember generation taskTypes (COLOR_GUIDED_GENERATION, TEXT_IMAGE) are invalid on /images/edits.
Use exact uppercase strings; casing is not normalized.

When it happens

Trigger: Passing taskType values valid for Nova text-to-image but not edits (e.g. 'COLOR_GUIDED_GENERATION', 'TEXT_IMAGE'), or casing/typo variants like 'inpainting', 'Inpainting', 'WATERMARK' in an images/edits call on a nova-canvas model.

Common situations: Copy-pasting a taskType from a Nova Canvas image-generation payload into an image-edit request; assuming the generation taskType enum applies to edits; sending whitespace-padded values is tolerated but wrong casing is not.

Related errors

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/f718b9fcd87f7048. Report an issue: GitHub.