sgl-project/sglang · error · RuntimeError

Failed to generate image: {str(e)}

Error message

Failed to generate image: {str(e)}

What it means

RuntimeError raised by the ComfyUI image node wrapping any exception from sgld_client.generate_image — network failures, server-side errors, or the upstream RuntimeErrors from server_api.py all surface here with their message chained.

Source

Thrown at python/sglang/multimodal_gen/apps/ComfyUI_SGLDiffusion/nodes.py:386

            request_params["num_inference_steps"] = steps
        if seed is not None and seed >= 0:
            request_params["seed"] = seed
        if enable_teacache:
            request_params["enable_teacache"] = True
        if image is not None:
            # If the image is empty, use the size of the image to generate the image
            if is_empty_image(image):
                width, height = image.shape[2], image.shape[1]
                size = f"{width}x{height}"
                request_params["size"] = size
            else:
                request_params["image_path"] = get_image_path(image)

        # Call API
        try:
            response = sgld_client.generate_image(**request_params)
        except Exception as e:
            raise RuntimeError(f"Failed to generate image: {str(e)}")

        # Decode base64 image
        if not response["data"] or not response["data"][0]["b64_json"]:
            raise RuntimeError("No image data in response")
        image_data = response["data"][0]["b64_json"]
        image = convert_b64_to_tensor_image(image_data)

        return (image,)


class SGLDiffusionGenerateVideo:
    """Node to generate videos using SGLang Diffusion."""

    @classmethod
    def INPUT_TYPES(cls):
        return {
            "required": {
                "sgld_client": ("SGLD_CLIENT",),

View on GitHub (pinned to 0132848349)

Solutions

1. Read the chained message — it contains the true underlying cause (timeout, status code, etc.).
2. Start/health-check the SGLang Diffusion server before running the workflow.
3. Fix the underlying cause per its own error guidance (auth, params, timeout).
4. Retry the workflow once the server is ready.

import requests
assert requests.get(f"{sgld_client.base_url}/models", timeout=5).ok

try:
    out = node.generate_image(...)
except RuntimeError as e:
    show_in_comfy_ui(str(e))  # chained message holds the real cause
    raise

When it happens

Trigger: Any failure of the underlying API call inside the node: server unreachable, generation error, or decode failure; the node re-raises as 'Failed to generate image: <cause>'.

Common situations: Server not running when the workflow executes; model still loading; bad api_key; generation parameters rejected by the server.

Understand the failure class

Background: 'Something went wrong' / 'Request failed (500)' / 'HTTP error! status: 404' — what failed HTTP requests actually mean and how to find the real cause — this error's family across 28 libraries.

Related errors

• Failed to generate video: {str(e)}
• Failed to generate image: {str(e)}
• Prompt cannot be empty
• No image data in response
• Failed to get server info. {error_data['error']['message']}

AI-assisted analysis of sgl-project/sglang@0132848349 (2026-08-28). Data as JSON: /api/errors/64dc4773ed79b06d. Report an issue: GitHub.