Unverified50% confidenceFactExact time
In the Firestore scenario, the retrieval step of RAG is usually implemented with vector search or ordinary queries filtered by field.
1
Sources
50%
Confidence
Long-term
Relevance
10/13/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn practice, RAG retrievers often combine multiple types of retrievers (similarity retrieval, range retrieval, grouped retrieval, hybrid retrieval, full-text retrieval, etc.).73% similarUnverifiedWeb search is a simplified implementation of RAG, in which the model directly calls a search engine to obtain real-time results.72% similarUnverifiedFireCrawl能把整个网站的内容完整抓取下来，可以设定爬取深度，与Tavily和Jina主要做搜索查询不同72% similarVerifiedIntent classification in RAG architecture routing mechanisms is sometimes referred to as Query Routing, and is one of the important optimization techniques for advancing from simple RAG to advanced RAG.71% similarUnverifiedExa搜索工具采用类似RAG的技术，预先将网站内容抓取并存储在向量数据库中，查询时使用模糊语义匹配71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1018092API
curl https://kongchang.com/api/v1/knowledge/claims/1018092MCP
get_claim(id=1018092)