Comfy-Org/ComfyUI · error · ValueError

The current maximum number of supported images is {OMNI_MAX_

Error message

The current maximum number of supported images is {OMNI_MAX_IMAGES}.

What it means

The omni (Interactions API) node enforces OMNI_MAX_IMAGES = 14: the summed image count over all connected image tensors may not exceed 14. Videos have their own separate cap. The check runs client-side before media parts are built, so no API quota is consumed when it fires.

Source

Thrown at comfy_api_nodes/nodes_gemini.py:1656

                IO.Hidden.api_key_comfy_org,
                IO.Hidden.unique_id,
            ],
            is_api_node=True,
            price_badge=IO.PriceBadge(
                expr='{"type":"usd","usd":0.101,"format":{"suffix":"/second","approximate":true}}'
            ),
        )

    @classmethod
    async def execute(cls, model: dict, seed: int) -> IO.NodeOutput:
        prompt = model.get("prompt") or ""
        validate_string(prompt, strip_whitespace=True, min_length=1)
        model_id = OMNI_MODELS[model["model"]]

        images = [t for t in (model.get("images") or {}).values() if t is not None]
        videos = [v for v in (model.get("videos") or {}).values() if v is not None]
        if sum(get_number_of_images(t) for t in images) > OMNI_MAX_IMAGES:
            raise ValueError(f"The current maximum number of supported images is {OMNI_MAX_IMAGES}.")
        if len(videos) > OMNI_MAX_VIDEOS:
            raise ValueError(f"The current maximum number of supported videos is {OMNI_MAX_VIDEOS}.")
        for video in videos:
            validate_video_duration(video, max_duration=10)

        parts: list[GeminiInteractionTextPart | GeminiInteractionMediaPart] = []
        if images or videos:
            # The Interactions API accepts video only inline or as a Files API URI, not as an HTTP URL.
            media_parts = await build_gemini_media_parts(
                cls, [], [], videos, url_budget=0, max_inline_bytes=GEMINI_INTERACTIONS_MAX_INLINE_BYTES
            )
            video_inline_bytes = sum(len(p.inlineData.data) for p in media_parts)
            media_parts += await build_gemini_media_parts(
                cls, images, [], [], max_inline_bytes=GEMINI_INTERACTIONS_MAX_INLINE_BYTES - video_inline_bytes
            )
            parts.extend(to_interaction_media_part(p) for p in media_parts)
        parts.append(GeminiInteractionTextPart(text=prompt))
        interaction = await sync_op(

View on GitHub (pinned to 1c6d8d45b3)

Solutions

Trim connected images so the total across all sockets is ≤14.
Move some visual context into the videos input (subject to OMNI_MAX_VIDEOS = 3) instead of many stills.
Split the task into multiple interactions with smaller image sets.

Example fix

# before
images = [batch_of_20_frames]
await omni_execute(model={'images': {'i1': images}, ...}, seed=s)  # raises

# after
images = [batch_of_20_frames[:14]]
await omni_execute(model={'images': {'i1': images}, ...}, seed=s)

Defensive patterns

Strategy: validation

Validate before calling

OMNI_MAX_IMAGES = 14
total = sum(get_number_of_images(t) for t in (model.get("images") or {}).values() if t is not None)
if total > OMNI_MAX_IMAGES:
    trim_images(model, limit=OMNI_MAX_IMAGES)

Prevention

Budget 14 images total across the omni node's image sockets.
Move some visual context into the (up to 3) video inputs instead.
Slice frame batches before wiring them in.

When it happens

Trigger: Executing the Gemini omni node with model['images'] tensors whose get_number_of_images sum exceeds OMNI_MAX_IMAGES (14). Batched tensors count per image.

Common situations: Feeding a long frame extraction plus reference photos into the omni node expecting it to handle video; porting workflows from the image-only nodes without recounting; batch sizes sized for other providers.

Related errors

AI-assisted analysis of Comfy-Org/ComfyUI@1c6d8d45b3 (2026-08-14). Data as JSON: /api/errors/830cce1cffbcd8c0. Report an issue: GitHub.