Valkey 是一个开源、高性能的键值数据存储，支持缓存、消息队列等工作负载，也可作为主数据库使用。Valkey 可以作为独立守护进程运行，也可以在集群中运行，并提供复制和高可用性选项。本页介绍如何将 Valkey 向量存储与 Amazon ElastiCache for Valkey 或 Amazon MemoryDB for Valkey 结合使用。
设置
安装所需的依赖项：pip install "langchain-aws[valkey]"
uv add langchain-aws --extra valkey
Valkey 集成需要
langchain-aws>=1.5.0。如果您使用的是早期版本，请直接安装依赖项：pip install langchain-aws valkey-glide-sync
基本用法
使用 Bedrock 嵌入
from langchain_aws import BedrockEmbeddings
from langchain_aws.vectorstores import ValkeyVectorStore
# 初始化嵌入
embeddings = BedrockEmbeddings(
model_id="amazon.titan-embed-text-v1",
region_name="us-east-1"
)
# 从文本创建向量存储
vectorstore = ValkeyVectorStore.from_texts(
texts=["Valkey is fast", "Valkey supports vector search"],
embedding=embeddings,
valkey_url="valkey://localhost:6379",
index_name="my_index"
)
# 执行相似性搜索
results = vectorstore.similarity_search("fast database", k=2)
for doc in results:
print(doc.page_content)
使用 Ollama 嵌入
from langchain_ollama import OllamaEmbeddings
from langchain_aws.vectorstores import ValkeyVectorStore
# 初始化 Ollama 嵌入
embeddings = OllamaEmbeddings(
model="nomic-embed-text",
base_url="http://localhost:11434"
)
# 创建向量存储
vectorstore = ValkeyVectorStore(
embedding=embeddings,
valkey_url="valkey://localhost:6379",
index_name="my_index",
vector_schema={
"name": "content_vector",
"algorithm": "FLAT",
"dims": 768, # nomic-embed-text 维度
"distance_metric": "COSINE",
"datatype": "FLOAT32",
}
)
# 添加文本
vectorstore.add_texts(
texts=["Document 1", "Document 2"],
metadatas=[{"source": "doc1"}, {"source": "doc2"}]
)
# 搜索
results = vectorstore.similarity_search("query", k=2)
连接 URL
ValkeyVectorStore 支持多种连接 URL 格式：# 独立模式
valkey_url = "valkey://localhost:6379"
# 带身份验证
valkey_url = "valkey://username:password@host:6379"
# SSL/TLS
valkey_url = "valkeyss://host:6379"
# 带身份验证的 SSL
valkey_url = "valkeyss://username:password@host:6379"
AWS ElastiCache for Valkey
from langchain_aws import BedrockEmbeddings
from langchain_aws.vectorstores import ValkeyVectorStore
embeddings = BedrockEmbeddings()
# 连接到 ElastiCache 集群
vectorstore = ValkeyVectorStore(
embedding=embeddings,
valkey_url="valkeyss://my-cluster.cache.amazonaws.com:6379",
index_name="my_index"
)
# 添加文档
vectorstore.add_texts(
texts=["Document 1", "Document 2"],
metadatas=[{"source": "doc1"}, {"source": "doc2"}]
)
元数据过滤
from langchain_aws.vectorstores.valkey.filters import ValkeyTag, ValkeyNum
# 添加带元数据的文档
vectorstore.add_texts(
texts=["AI article from 2024", "ML paper from 2023"],
metadatas=[
{"category": "ai", "year": 2024},
{"category": "ml", "year": 2023}
]
)
# 使用过滤器搜索
filter_expr = (ValkeyTag("category") == "ai") & (ValkeyNum("year") >= 2024)
results = vectorstore.similarity_search(
"artificial intelligence",
k=5,
filter=str(filter_expr)
)
自定义向量模式
from langchain_aws.vectorstores import ValkeyVectorStore
vectorstore = ValkeyVectorStore(
embedding=embeddings,
valkey_url="valkey://localhost:6379",
index_name="my_index",
vector_schema={
"name": "content_vector",
"algorithm": "HNSW", # 或 "FLAT"
"dims": 1536,
"distance_metric": "COSINE", # 或 "L2", "IP"
"datatype": "FLOAT32",
}
)
API 参考
有关详细的 API 文档，请参阅ValkeyVectorStore。
将这些文档连接到 Claude、VSCode 等，通过 MCP 获取实时答案。