Qdrant 的 Nano 与 Zero 可与 Stella 文档向量互换使用
做检索的朋友注意:Qdrant 的 Nano 和 Zero 查询向量能直接配 Stella 的文档索引,延迟和质量可以按需切换,不用重建索引。
Qdrant 介绍了 Nano 与 Zero 两个向量模型的兼容设计:两者都针对 Stella 的文档嵌入向量训练,生成的查询向量可直接匹配同一份文档索引。Zero 极其轻量,Nano 是小型 transformer,检索质量更接近 Stella。用户可按质量与查询延迟的权衡在两者间切换,无需重建文档索引。
/4 Nano and Zero are specifically trained to produce vectors compatible with Stella’s document embeddings. So you can switch between them depending on the trade-off you want between quality and query latency, without rebuilding the document index. For example, Zero is extremely lightweight, while Nano is a small transformer that gets much closer to Stella’s retrieval quality. 💬 0 🔄 0 ❤️ 0 👀 13 ⚡