[Feature]: How can I get embedding result from images？ Can Qwen2.5-vl-7b do this?

[LwengGitHub]

🚀 The feature, motivation and pitch

URL:XXXXXXX:8000/v1/embeddings
###############################
{
"model": "qwen2.5-vl-7b",
"encoding_format": "float",
"messages": [
{
"role": "user",
"content": [
{
"type": "image_url",
"image_url": {
"url": "data:image/png;base64,[base64_info]"
}
}
]
}
]
}

Alternatives

No response

Additional context

No response

Before submitting a new issue...

• Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

[DarkLight1337]

I assume you are talking about getting the image embeddings (i.e. output of visual encoder) without running the language backbone. No, this is not supported yet.