Cohere 发布 Embed 5 系列前沿嵌入模型
原标题：Introducing Embed 5—A New Family of Frontier Embedding Models
AI 摘要
Cohere 发布 Embed 5 系列嵌入模型，包含面向最高检索质量的 Embed 5 Pro 和面向低延迟低成本场景的 Embed 5 Fast。两款模型共享同一嵌入空间，支持 128K 上下文、100+ 语言、多模态输入与 Matryoshka 嵌入，并已在 Cohere API、Model Vault、Microsoft Foundry 和 Amazon SageMaker 上线。Cohere 称 Embed 5 Pro 在 ViDoRe V3、金融文档、解析 PDF、图像检索等基准上取得领先成绩，定价为 Pro 每百万 token 0.12 美元、Fast 每百万 token 0.08 美元。
正文节选
Today, we're releasing Embed 5, a new family of embeddings models at the frontier of high-quality enterprise retrieval. Embed 5 delivers stronger retrieval across complex enterprise data while giving teams more control over latency, cost, and deployment. Embed 5 Pro is optimized for maximum quality across multimodal, multilingual, financial, code, and parsed-document retrieval. Embed 5 Fast brings highly competitive performance to latency- and cost-sensitive workloads. Both tiers share a single