calesthio/OpenMontage · error · ValueError

prompt exceeds Kling image generation limit of 2500 characte

Error message

prompt exceeds Kling image generation limit of 2500 characters

What it means

ValueError raised when the prompt for a Kling image generation request exceeds 2500 characters, the documented API limit. The check runs client-side during request building, before any network call, so no quota is consumed.

Source

Thrown at tools/graphics/kling_official_image.py:268

            api_family = "omni"
        if api_family == "omni":
            return self._build_omni_request(inputs)
        return self._build_generation_request(inputs)

    def _build_generation_request(self, inputs: dict[str, Any]) -> dict[str, Any]:
        prompt = self._prompt(inputs)
        model_name = str(inputs.get("model_name") or "kling-v3")
        if model_name not in IMAGE_GENERATION_MODELS:
            raise ValueError(f"model_name {model_name!r} is not supported for api_family=generation")
        payload: dict[str, Any] = {
            "model_name": model_name,
            "prompt": prompt,
            "resolution": inputs.get("resolution", "1k"),
            "n": int(inputs.get("n", 1) or 1),
            "aspect_ratio": inputs.get("aspect_ratio", "16:9"),
        }
        if len(prompt) > 2500:
            raise ValueError("prompt exceeds Kling image generation limit of 2500 characters")
        if inputs.get("negative_prompt"):
            payload["negative_prompt"] = inputs["negative_prompt"]
        image = normalize_image_input(inputs.get("image_url"), inputs.get("image_path"))
        if image:
            payload["image"] = image
        if inputs.get("image_reference"):
            payload["image_reference"] = inputs["image_reference"]
        for key in ("image_fidelity", "human_fidelity"):
            if inputs.get(key) is not None:
                payload[key] = inputs[key]
        elements = normalize_element_list(inputs.get("element_list"))
        if elements:
            payload["element_list"] = elements
        self._copy_common_task_fields(inputs, payload)
        return {
            "protocol": "classic",
            "path": "/v1/images/generations",
            "payload": payload,

View on GitHub (pinned to 95e1c3d0ab)

Solutions

1. Trim the prompt to under 2500 characters; move secondary detail into negative_prompt (subject to its own limits)
2. Compress verbose descriptions to the essential subject, style, and composition cues
3. If using an agent, add a truncation/summarization step before the tool call
4. Compute len(prompt) before invoking the tool to fail fast in your own code

Example fix

// before
inputs = {"prompt": very_long_description}  // 4000 chars
// after
prompt = very_long_description[:2490].rsplit(" ", 1)[0]
inputs = {"prompt": prompt}

assert len(prompt) <= 2500, f"prompt is {len(prompt)} chars; Kling limit is 2500"

When it happens

Trigger: Passing a very long prompt (elaborate scene descriptions, pasted context, or an agent concatenating instructions) to the generation API family; prompts padded with references or negative instructions.

Common situations: LLM agents generating unbounded prompt text; concatenating prompt + style descriptors + camera notes into one string; localizing to character-heavy languages where 2500 chars arrive quickly.

Related errors

• model_name {model_name!r} is not supported for api_family=ge
• model_name {model_name!r} is not supported for api_family=om
• image_list items must be objects
• image_list items must include image, image_url, or image_pat
• element_list must be a list of element ids or objects

AI-assisted analysis of calesthio/OpenMontage@95e1c3d0ab (2026-08-15). Data as JSON: /api/errors/38480f99e132b422. Report an issue: GitHub.