I would try the Qwen models before LLaVa Do you need the embeddings to be privat... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		stormfather 5 months ago \| parent \| context \| favorite \| on: Ask HN: What's the 2025 stack for a self-hosted ph... I would try the Qwen models before LLaVa Do you need the embeddings to be private? Or just the photos?

msgodel 5 months ago [–]

For photo indexing I'd run CLIP directly and save on compute, no need to use a whole language model.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact