Skip to main content

​
The problem

Keyword search misses results that use different words for the same idea. “How do I get my money back” won’t match a page titled “Refund policy.”

​
How Ragrails solves it

Embeddings match by meaning. Add reranking to push the most relevant results to the top.
  1. Ingest your corpus.
  2. Retrieve by meaning, retrieving wide.
  3. Rerank down to the few best results.
from ragrails import RagRails

rag = RagRails()
embedder = rag.embedder(provider="voyage", model="voyage-3", input_type="query")
reranker = rag.reranker(provider="voyage", model="rerank-2-lite")

result = rag.retrieve(
    "how do I get my money back",
    embedder=embedder,
    vector_db="qdrant", collection="docs", url="http://localhost:6333",
    top_k=20,
    use_rerank=True,
    reranker=reranker,
    rerank_top_k=5,
)
for hit in result.items:
    print(hit.score, hit.metadata.get("title"), hit.text[:120])
No answer generation and no LLM cost; this is pure retrieval. Add chat on top if you want generated answers.

​
Features used

Retrieval · Reranking · Embedding