- Announces GA of Amazon Nova Multimodal Embeddings in AWS GovCloud (US‑West) as a unified model for text, images, video, and audio.
- Supports up to 8K tokens and 30‑second audio/video segments, with synchronous and asynchronous APIs and configurable output dimensions for accuracy‑performance trade‑offs.
- Enables cross‑modal retrieval, simplifying multi‑modal search and reducing cost and data silos.