invoke-ai/InvokeAI · error · ValueError

At least one Krea-2 text conditioning is required.

Error message

At least one Krea-2 text conditioning is required.

What it means

Krea2RegionalPromptingExtension.from_text_conditionings assembles concatenated embeddings and embedding ranges from the supplied text conditionings. With an empty list there is nothing to build a regional layout from, so it raises ValueError immediately.

Source

Thrown at invokeai/backend/krea2/regional_prompting.py:54

    def attention_mask_numel(self) -> int:
        if not self.has_regional_masks:
            return 0
        total_seq_len = self.regional_text_conditioning.prompt_embeds.shape[1] + self.image_seq_len
        return total_seq_len**2

    @property
    def attention_mask_build_scratch_numel(self) -> int:
        """Peak boolean scratch allocation used while constructing the image-to-image attention block."""
        if not self.has_regional_masks:
            return 0
        return self.image_seq_len**2

    @classmethod
    def from_text_conditionings(
        cls, text_conditionings: list[Krea2TextConditioning], image_seq_len: int
    ) -> "Krea2RegionalPromptingExtension":
        if not text_conditionings:
            raise ValueError("At least one Krea-2 text conditioning is required.")

        prompt_embeds: list[torch.Tensor] = []
        image_masks: list[torch.Tensor | None] = []
        embedding_ranges: list[Range] = []
        current_start = 0
        for conditioning in text_conditionings:
            sequence_length = conditioning.prompt_embeds.shape[1]
            if conditioning.mask is not None and conditioning.mask.numel() != image_seq_len:
                raise ValueError(
                    f"Krea-2 regional mask has {conditioning.mask.numel()} values, expected {image_seq_len}."
                )
            prompt_embeds.append(conditioning.prompt_embeds)
            image_masks.append(conditioning.mask)
            embedding_ranges.append(Range(start=current_start, end=current_start + sequence_length))
            current_start += sequence_length

        regional_text_conditioning = Krea2RegionalTextConditioning(
            prompt_embeds=torch.cat(prompt_embeds, dim=1),

View on GitHub (pinned to 0b6a024f2f)

Solutions

1. Ensure at least one Krea2TextConditioning (typically the global/base prompt) is passed.
2. Check upstream code that filters or splits prompts into regional conditionings and fix the empty-list case.
3. Disable regional prompting if you only have a single non-regional prompt.
4. Assert len(text_conditionings) > 0 before calling the classmethod.

Example fix

// before
ext = Krea2RegionalPromptingExtension.from_text_conditionings(regions, image_seq_len)  # regions == []
// after
if not regions:
    regions = [default_text_conditioning]
ext = Krea2RegionalPromptingExtension.from_text_conditionings(regions, image_seq_len)

if not text_conditionings:
    raise ValueError('Regional prompting enabled but no text conditionings provided; add at least the base prompt')
ext = Krea2RegionalPromptingExtension.from_text_conditionings(text_conditionings, image_seq_len)

def can_build_regional_extension(conds) -> bool:
    return isinstance(conds, (list, tuple)) and len(conds) > 0

try:
    ext = Krea2RegionalPromptingExtension.from_text_conditionings(conds, image_seq_len)
except ValueError as e:
    if 'At least one Krea-2 text conditioning' in str(e):
        ext = None  # fall back to non-regional generation
    else:
        raise

When it happens

Trigger: Calling from_text_conditionings([], image_seq_len=N) — e.g. the regional-prompting path was enabled but regional prompt extraction produced zero conditionings, or a filter dropped all conditionings before this call.

Common situations: Regional prompting toggle enabled with no regional prompts defined; a bug upstream that skips adding the base conditioning; passing an empty slice of conditionings.

Related errors

• per_layer_weights must be comma-separated numbers: {e}
• per_layer_weights must have exactly {_NUM_TEXT_LAYERS} value
• per_layer_weights must contain only finite values.
• 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/de57559fba717fd1. Report an issue: GitHub.