microsoft/semantic-kernel · error · ServiceInvalidExecutionSettingsError

Image Content URI needs to be set, because onnx can only wor

Error message

Image Content URI needs to be set, because onnx can only work with file paths

What it means

Raised inside _get_images_from_history when the model IS multi-modal (enable_multi_modality is True) but an ImageContent item has a falsy .uri attribute. The ONNX runtime's Pybind layer can only load images from file paths (OnnxRuntimeGenAi.Images.open), not from base64 data or byte buffers. Raised as ServiceInvalidExecutionSettingsError.

Source

Thrown at python/semantic_kernel/connectors/ai/onnx/services/onnx_gen_ai_chat_completion.py:195

    def _chat_messages_to_dicts(self, chat_history: "ChatHistory") -> list[dict[str, Any]]:
        return [
            message.to_dict(role_key="role", content_key="content")
            for message in chat_history.messages
            if isinstance(message, ChatMessageContent)
        ]

    def _get_images_from_history(self, chat_history: "ChatHistory") -> list[ImageContent] | None:
        images = []
        for message in chat_history.messages:
            for image in message.items:
                if isinstance(image, ImageContent):
                    if not self.enable_multi_modality:
                        raise ServiceInvalidExecutionSettingsError("The model does not support multi-modality")
                    if image.uri:
                        images.append(image)
                    else:
                        raise ServiceInvalidExecutionSettingsError(
                            "Image Content URI needs to be set, because onnx can only work with file paths"
                        )
        return images if len(images) else None

    def _get_audios_from_history(self, chat_history: "ChatHistory") -> list[AudioContent] | None:
        audios = []
        for message in chat_history.messages:
            for audio in message.items:
                if isinstance(audio, AudioContent):
                    if not self.enable_multi_modality:
                        raise ServiceInvalidExecutionSettingsError("The model does not support multi-modality")
                    if audio.uri:
                        audios.append(audio)
                    else:
                        raise ServiceInvalidExecutionSettingsError(
                            "Audio Content URI needs to be set, because onnx can only work with file paths"
                        )
        return audios if len(audios) else None

View on GitHub (pinned to c028a0c7dc)

Solutions

Set image.uri to a local file path when using ONNX vision models
If you have base64 or bytes, write to a temp file and set uri to that path
Use ImageContent(uri='file:///path/to/image.png') instead of data-only construction

Example fix

// before
ImageContent(data=base64_encoded_string)
// after
ImageContent(uri='file:///tmp/image.png')

Defensive patterns

Strategy: validation

Validate before calling

from semantic_kernel.contents import ImageContent

def validate_image_uris(chat_history):
    for msg in chat_history.messages:
        for item in msg.items:
            if isinstance(item, ImageContent) and not item.uri:
                raise ValueError('ImageContent must have a uri for ONNX models')

Type guard

from semantic_kernel.contents import ImageContent

def all_images_have_uri(chat_history) -> bool:
    return all(
        item.uri for msg in chat_history.messages
        for item in msg.items if isinstance(item, ImageContent)
    )

Try / catch

from semantic_kernel.exceptions.service_exceptions import ServiceInvalidExecutionSettingsError

try:
    result = await chat.get_chat_message_contents(chat_history=history, settings=settings)
except ServiceInvalidExecutionSettingsError as e:
    if 'Image Content URI needs to be set' in str(e):
        for msg in history.messages:
            for item in msg.items:
                if isinstance(item, ImageContent) and not item.uri:
                    item.uri = 'file:///tmp/resolved_image.png'
        result = await chat.get_chat_message_contents(chat_history=history, settings=settings)

Prevention

Always set ImageContent.uri to a local file path for ONNX vision models
Write base64 image data to a temp file and use its path if only data is available
Pre-validate all image URIs exist on disk before calling the service

When it happens

Trigger: Adding ImageContent(data='base64...') to chat history for a vision ONNX model without setting the uri. The formatter checks image.uri and rejects None/empty.

Common situations: Creating ImageContent with only base64 data (as required by Ollama) and reusing it with ONNX; loading images from a URL or byte stream without writing to a temp file; ONNX's limitation of file-path-only image loading not accounted for.

Related errors

AI-assisted analysis of microsoft/semantic-kernel@c028a0c7dc (2026-08-13). Data as JSON: /api/errors/b3f33f3fce416d92. Report an issue: GitHub.