July 08, 2025 — Snowflake AISQL AI_EMBED multimodal embeddings — Preview¶
AISQL AI_EMBED is now available in preview, enabling customers to generate high-quality image and text embedding vectors directly within Snowflake using simple SQL. Embeding vectors allow text and images to be compared and search based on their features.
AI_EMBED allows organizations to:
Build sophisticated image search and similarity systems: Find visually similar products, medical images, or design assets across massive datasets.
Convert complex documents and images into searchable vector representations: Transform unstructured visual content into queryable data.
Enhance content moderation workflows: Automatically detect and flag inappropriate visual content across user-generated media.
Streamline digital asset management: Organize and retrieve marketing materials, brand assets, and creative content through semantic image search.
Support manufacturing quality control: Identify defects and anomalies in product images by comparing against reference standards.
Enable intelligent document processing: Extract insights from invoices, contracts, and forms by embedding both text and visual layout information.
To learn more, see Cortex AISQL Images and AI_EMBED.