Jina Embeddings v5 Omni Nano by Jina AI — 768D

1.04B-parameter compact multimodal omni embedding model accepting text, images, video, and audio. Shared vector space with text-only v5-text-nano. 768-dim embeddings with Matryoshka truncation. Supports retrieval, classification, clustering, and text-matching tasks. Optimized for efficiency.