invoke-ai/InvokeAI · error · ValueError

denoising_start should be 0 when initial latents are not pro

Error message

denoising_start should be 0 when initial latents are not provided.

What it means

When no initial latents (init_latents) are supplied, the denoise run starts from pure noise, which corresponds to sigma step 0. If denoising_start > 1e-5 the schedule would pretend some steps already ran, so InvokeAI raises this ValueError.

Source

Thrown at invokeai/app/invocations/anima_denoise.py:628

        # Prepare input latents
        if init_latents is not None:
            if self.add_noise:
                assert noise is not None
                # Noise the init latents using the first sigma from the clipped
                # InvokeAI schedule.
                #
                # Known limitation: if the selected scheduler later starts from a
                # different first effective sigma/timestep than sigmas[0], the
                # img2img preblend below may not match that scheduler exactly.
                # This is an existing pipeline limitation and affects both
                # internally generated noise and externally supplied noise.
                s_0 = sigmas[0]
                latents = s_0 * noise + (1.0 - s_0) * init_latents
            else:
                latents = init_latents
        else:
            if self.denoising_start > 1e-5:
                raise ValueError("denoising_start should be 0 when initial latents are not provided.")
            assert noise is not None
            latents = noise

        if total_steps <= 0:
            return latents.squeeze(2)

        # Prepare inpaint extension
        inpaint_mask = self._prep_inpaint_mask(context, latents.squeeze(2))
        inpaint_extension: AnimaInpaintExtension | None = None
        if inpaint_mask is not None:
            if init_latents is None:
                raise ValueError("Initial latents are required when using an inpaint mask (image-to-image inpainting)")
            assert noise is not None
            inpaint_extension = AnimaInpaintExtension(
                init_latents=init_latents.squeeze(2),
                inpaint_mask=inpaint_mask,
                noise=noise.squeeze(2),
                shift=ANIMA_SHIFT,

View on GitHub (pinned to 0b6a024f2f)

Solutions

1. Set denoising_start to 0 when running without initial latents.
2. Connect an init image/latents input if you intend to do img2img with a nonzero denoising_start.
3. Validate the combination in workflow-building code before invoking.

Example fix

// before
denoise.denoising_start = 0.4  # no init latents connected
// after
denoise.denoising_start = 0.0  # or provide init_latents

if init_latents is None and denoise.denoising_start > 1e-5:
    denoise.denoising_start = 0.0  # or supply init latents

try:
    output = invoker.invoke(denoise_invocation)
except ValueError as e:
    if "initial latents are not provided" in str(e):
        denoise.denoising_start = 0.0
        output = invoker.invoke(denoise_invocation)
    else:
        raise

When it happens

Trigger: Running Anima denoise in pure txt2img mode (no init_latents input connected) while denoising_start is set to a positive value greater than 1e-5.

Common situations: A txt2img workflow where a leftover img2img denoise fraction was left on the node; an img2img image input got disconnected so init_latents became None but the start value stayed; script automation that sets denoising_start without checking whether init image is provided.

Related errors

• denoising_start ({self.denoising_start}) must be less than d
• Initial latents are required when using an inpaint mask (ima
• denoising_start must be 0 when no initial latents are provid
• denoising_start should be 0 when initial latents are not pro
• denoising_start should be 0 when initial latents are not pro

AI-assisted analysis of invoke-ai/InvokeAI@0b6a024f2f (2026-08-29). Data as JSON: /api/errors/135806a40665e3e6. Report an issue: GitHub.