`JinaDocumentImageEmbedder`

[anakin87]

Some Jina models support image inputs: jina-clip series + jina-embeddings-v4.

It would be nice to have a component that supports them via their API.
Refer to SentenceTransformersDocumentImageEmbedder for the implementation.

Relevant info:

• https://jina.ai/models/jina-embeddings-v4
• https://jina.ai/api-dashboard/embedding

---

Tasks

• The code is documented with docstrings and was merged in the main branch
• The package has been released on PyPI
• The integration has been listed in the Inventory section of this repo README
• Updated Jina documentation, add DocumentImageEmbedder haystack-integrations#354
• Publish JinaDocumentImageEmbedder docs #2244

[anakin87]

@ChinmayBansal since this is a new component, we should include a small example in http://github.com/deepset-ai/haystack-integrations/blob/main/integrations/jina.md (source for https://haystack.deepset.ai/integrations/jina)

LMK if you would like to open a PR to update that file. In case you want, feel free to update the content, which might be a bit outdated.

[ChinmayBansal]

I'd be happy to do update the documentation to include the new component. I will add an image embedding example, updated component list, an example, and update an outdated information I find. Should I create this is a separate PR for documentation updates or add it to the current PR?

[anakin87]

Thanks! It's a separate repo, so you would need a different PR.