by Perplexity
Open-weights vision-language embedding models from Perplexity (9B and 600M) for text, image and document retrieval.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
Perplexity Embed V2 Context is a family of vision-language embedding models released by Perplexity with open weights, in 9B and 600M parameter sizes. According to a creator summary, they are designed for retrieval over text, images and visual documents such as PDFs, and a paper was published alongside the release.
The models map text and visual content into a shared embedding space, so they can be used for search and retrieval-augmented generation pipelines that include scanned or image-heavy documents. The two sizes let users trade accuracy against compute cost.
The provided pages did not include a model card, license, benchmark figures, or exact release date, so those details are not stated here. Check the vendor's documentation and the Hugging Face release for specifics.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
Open-weights vision-language embedding models in 9B and 600M sizes, for retrieval over text, images and visual documents such as PDFs. A paper was released alongside it.
Closest neighbours by meaning, not just by category.
Model · Perplexity
Two open-sourced Perplexity decision models, at 4B and 27B parameters, announced by the company.
Model · AIDC-AI
Open multimodal embedding models from AIDC-AI that map text, images, video and documents into a shared vector space.
Model · Google
Google's small open embedding model that maps text, code, images, video and audio into a shared embedding space.
Model · NVIDIA
NVIDIA family of multilingual text embedding models (8B and 1B) for retrieval, RAG and coding agents.
Model · Princeton
Fully open 3B text-to-image diffusion model from Princeton with released weights, training code, and data pipeline.
Model · SenseTime
9B-parameter looped multimodal model from SenseTime that reuses a layer block twice for long-context and agentic tasks.