invoke-ai/InvokeAI · error · ValueError

{self.class} has the config param `addition_embed_type`

Error message

{self.__class__} has the config param `addition_embed_type` set to 'text_time' which requires the keyword argument `text_embeds` to be passed in `added_cond_kwargs`

What it means

For models with addition_embed_type 'text_time' (SDXL-style UNets), forward() needs micro-conditioning text embeds supplied via added_cond_kwargs['text_embeds']. Missing it means the additive embedding branch cannot be computed, so this patched forward raises immediately.

Source

Thrown at invokeai/backend/util/hotfixes.py:687

        aug_emb = None

        if self.class_embedding is not None:
            if class_labels is None:
                raise ValueError("class_labels should be provided when num_class_embeds > 0")

            if self.config.class_embed_type == "timestep":
                class_labels = self.time_proj(class_labels)

            class_emb = self.class_embedding(class_labels).to(dtype=self.dtype)
            emb = emb + class_emb

        if "addition_embed_type" in self.config:
            if self.config.addition_embed_type == "text":
                aug_emb = self.add_embedding(encoder_hidden_states)

            elif self.config.addition_embed_type == "text_time":
                if "text_embeds" not in added_cond_kwargs:
                    raise ValueError(
                        f"{self.__class__} has the config param `addition_embed_type` set to 'text_time' which \
                            requires the keyword argument `text_embeds` to be passed in `added_cond_kwargs`"
                    )
                text_embeds = added_cond_kwargs.get("text_embeds")
                if "time_ids" not in added_cond_kwargs:
                    raise ValueError(
                        f"{self.__class__} has the config param `addition_embed_type` set to 'text_time' which \
                            requires the keyword argument `time_ids` to be passed in `added_cond_kwargs`"
                    )
                time_ids = added_cond_kwargs.get("time_ids")
                time_embeds = self.add_time_proj(time_ids.flatten())
                time_embeds = time_embeds.reshape((text_embeds.shape[0], -1))

                add_embeds = torch.concat([text_embeds, time_embeds], dim=-1)
                add_embeds = add_embeds.to(emb.dtype)
                aug_emb = self.add_embedding(add_embeds)

        emb = emb + aug_emb if aug_emb is not None else emb

View on GitHub (pinned to 0b6a024f2f)

Solutions

Pass added_cond_kwargs={"text_embeds": pooled_prompt_embeds, "time_ids": ...} computed from the text encoder's pooled output
Use the standard SDXL pipeline (StableDiffusionXLPipeline) which populates added_cond_kwargs for you
Check that the checkpoint/config actually matches the call signature you are using

Example fix

// before
unet(latents, t, encoder_hidden_states)
// after
unet(latents, t, encoder_hidden_states, added_cond_kwargs={"text_embeds": pooled_embeds, "time_ids": time_ids})

Defensive patterns

Strategy: validation

Validate before calling

if getattr(unet.config, 'addition_embed_type', None) == 'text_time':
    added_cond_kwargs = added_cond_kwargs or {}
    assert 'text_embeds' in added_cond_kwargs, "SDXL UNet needs text_embeds"

Type guard

def has_text_embeds(unet, added_cond_kwargs):
    if getattr(unet.config, 'addition_embed_type', None) != 'text_time':
        return True
    return isinstance(added_cond_kwargs, dict) and 'text_embeds' in added_cond_kwargs

Try / catch

try:
    out = unet(x, t, emb, added_cond_kwargs=added_cond_kwargs)
except ValueError as e:
    logger.error("missing added_cond_kwargs: %s", e)
    raise

Prevention

Always build added_cond_kwargs from the pooled text encoder output for SDXL
Keep text_embeds and time_ids constructed together
Use StableDiffusionXLPipeline instead of manual forward where possible

When it happens

Trigger: Calling forward() on an SDXL UNet (addition_embed_type='text_time') without added_cond_kwargs, or with added_cond_kwargs lacking the 'text_embeds' key; running a base-UNet call path that only passes time_ids.

Common situations: Custom SDXL sampling/LoRA code that forgot text_embeds; reusing SD 1.5-style forward calls against an SDXL checkpoint; pipelines that strip added_cond_kwargs.

Understand the failure class

Background: "missing required argument" and "the following required arguments were not provided": what required-argument errors mean and how to fix them — this error's family across 20 libraries.

Related errors

AI-assisted analysis of invoke-ai/InvokeAI@0b6a024f2f (2026-08-29). Data as JSON: /api/errors/fba87c3eeb6ffcde. Report an issue: GitHub.