Perplexity has released pplx-embed-v2-late, a pair of ColBERT-style multimodal embedding models. They come in two sizes: 0.6B for fast, cost-efficient queries, and 9B for maximum quality. Both models retrieve text, images, and rendered PDF pages, and they share a single embedding space.
Availability and Deployment
Deployable? Yes, if you self-host. Both models are available on Hugging Face under the MIT license. A hosted Perplexity API endpoint is planned but not yet live.
Quick Overview
[Content continues with detailed benchmarks, architecture notes, and comparison tables as per the original article.]
via MarkTechPost