Sentence Transformers — Multimodal Embedding & Reranker Training
Sentence Transformers now handles multimodal embeddings and rerankers end-to-end, letting you train models that work across text, images, and other modalities on standard hardware. You get production-ready architectures with minimal boilerplate—no need to stitch together separate libraries. Solid for teams building search, retrieval, or ranking systems that handle mixed content types.