详细内容
Meet us at re:Invent 2025. Come by Booth #534 and get a demo - Learn more Dismiss
Build Knowledgeable AI
The vector database for scale in production
{rag}
{search}
{recommendations}
{agents}
Performance at scale for {rag}
The purpose-built vector database delivering relevant results at any scale.
Customer workload: Popular productivity app providing instant Q&A across company knowledge
7.5B+
7.5B+
Total vectors
1.5M
1.5M
Namespaces
30M
30M
Global writes per day
Trusted in production
The world's most innovative companies are already in production with Pinecone.
USE CASE: RECOMMENDATIONS\
\
Read case study\
\
Gong achieves efficient vector searches, empowering Smart Trackers to offer users precise and relevant examples for concept tracking in conversations. USE CASE: SEARCH\
\
\
Read case study\
\
Before Pinecone, Vanguard’s customer support teams relied on keyword-based search solutions to search for documents where answers to a customer’s question may live. With Pinecone and hybrid retrieval, they boosted customer support with faster calls and 12% more accurate responses.
USE CASE: AGENTS
"Pinecone also supports hybrid search, combining sparse and dense embeddings, to deliver a more robust and accurate search experience. This flexibility allows us to optimize costs and performance, whether dealing with enterprises with extensive documentation or smaller companies with fewer pages."
USE CASE: RAG
"Pinecone aligns with our vision to democratize data accessibility for all engineers, and we're excited to uncover more potential with its new serverless architecture."
Developer Experience
Scale simplified
Fully managed and serverless for effortless scaling.
Rapid setup
Launch your vector databases in seconds.
Serverless scaling
Resources adjust to meet your demand automatically.
Rock-solid reliability
Trust in consistent uptime for your critical applications.
agent/retriever.py
Search
Relevance, delivered
Advanced retrieval capabilities for precise search across dynamic datasets.
Embeddings \
\
Choose from our leading hosted models or bring your own vectors. Optimized recall \
\
Benchmark leading algorithms maximize recall with low latency. Filters \
\
Retrieve only the vectors that match your metadata filters. Real-time indexing \
\
Upserted and updated vectors are dynamically indexed in real-time to ensure fresh reads. Full-text search \
\
Get an exact keyword match with sparse indexes when semantic search isn't enough.
Add an extra layer of precision with rerankers to boost the most relevant matches.
Create partitions of your data with namespaces to ensure tenant isolation.
Learn how to achieve best-in-class relevance with cascading retrieval
Works where you do
Use Pinecone with your favorite cloud provider, data sources, models, frameworks, and more.
Enterprise-ready AI
Meet security and operational requirements to bring AI products to market faster.
Secure
With encryption at rest and in transit, hierarchical encryption keys, private networking, and more, your data is secure. Contact us to deploy a privately managed Pinecone region within your cloud.
Reliable
Powering mission-critical applications of all sizes, with uptime SLAs, support SLAs, and observability.
Compliant
Control your data and know it's safe. Pinecone is SOC 2, GDPR, ISO 27001, and HIPAA certified.
Start building knowledgeable AI today
Create your first index for free, then pay as you go when you're ready to scale.
Subscribe to Pinecone
Subscribe