【2026年最新】RAG（検索拡張生成）実装ガイド：基礎から本番運用まで完全解説

from langchain.text_splitter import RecursiveCharacterTextSplitter

splitter = RecursiveCharacterTextSplitter(
    chunk_size=512,
    chunk_overlap=64,
    separators=["\n\n", "\n", "。", "、", " "],
    length_function=len,
)

chunks = splitter.split_documents(documents)

from langchain_experimental.text_splitter import SemanticChunker
from langchain_openai import OpenAIEmbeddings

embeddings = OpenAIEmbeddings(model="text-embedding-3-large")
semantic_splitter = SemanticChunker(
    embeddings,
    breakpoint_threshold_type="percentile",
    breakpoint_threshold_amount=90,
)

semantic_chunks = semantic_splitter.split_documents(documents)

from langchain_openai import OpenAIEmbeddings
from langchain_community.vectorstores import Qdrant

embeddings = OpenAIEmbeddings(model="text-embedding-3-large")

vectorstore = Qdrant.from_documents(
    documents=chunks,
    embedding=embeddings,
    url="http://localhost:6333",
    collection_name="knowledge_base",
    force_recreate=True,
)

from langchain_openai import ChatOpenAI
from langchain.chains import RetrievalQA
from langchain.prompts import PromptTemplate

llm = ChatOpenAI(model="gpt-4o", temperature=0)

prompt_template = """以下のコンテキストを参考に、ユーザーの質問に正確に回答してください。
コンテキストに情報がない場合は「情報が見つかりません」と回答してください。

コンテキスト:
{context}

質問: {question}

回答:"""

PROMPT = PromptTemplate(
    template=prompt_template,
    input_variables=["context", "question"]
)

qa_chain = RetrievalQA.from_chain_type(
    llm=llm,
    chain_type="stuff",
    retriever=vectorstore.as_retriever(search_kwargs={"k": 5}),
    chain_type_kwargs={"prompt": PROMPT},
    return_source_documents=True,
)

result = qa_chain.invoke({"query": "RAGの主なメリットは何ですか？"})

from langchain.retrievers import EnsembleRetriever
from langchain_community.retrievers import BM25Retriever

bm25_retriever = BM25Retriever.from_documents(chunks)
bm25_retriever.k = 5

vector_retriever = vectorstore.as_retriever(search_kwargs={"k": 5})

ensemble_retriever = EnsembleRetriever(
    retrievers=[bm25_retriever, vector_retriever],
    weights=[0.3, 0.7]
)

from langchain.retrievers import ContextualCompressionRetriever
from langchain_cohere import CohereRerank

reranker = CohereRerank(model="rerank-v3.5", top_n=3)

compression_retriever = ContextualCompressionRetriever(
    base_compressor=reranker,
    base_retriever=ensemble_retriever,
)

retriever = vectorstore.as_retriever(
    search_kwargs={
        "k": 5,
        "filter": {
            "department": "engineering",
            "year": {"$gte": 2025}
        }
    }
)