Skip to main content

Decision brief

Pinecone Review: Managed Vector Search and What It Costs

Pinecone serves hybrid search with hosted embedding and reranking on managed infrastructure, past US$50 and US$500 monthly minimums. Price your workload in read and write units, and pin your SDKs.

Reading time
7 min read
Sources
36 cited

GILD verdict

Choose Pinecone when your team wants hosted vector search with embedding and reranking built in, in Pinecone's regions or your own cloud on Enterprise.

Price the workload in read units, write units and storage at Standard rates before you commit, since the US$50 monthly minimum and usage past it decide the bill. If your data already lives in Postgres, pgvector saves a second database.

Pinecone

Vector Database by Pinecone Systems

−63 Strong negative
Product page

Fit and limits

Who it fits

  • You want no vector infrastructure to run: Pinecone runs on fully managed infrastructure on AWS, GCP and Azure.
  • You want embedding and reranking in the same service: Pinecone hosts embedding models, defaulting to llama-text-embed-v2 for dense vectors, or accepts your own vectors up to 20,000 dimensions.
  • Multi-tenant data: namespaces keep each tenant's data separate within one index.
  • Enterprise controls: SSO on Standard and Enterprise, SCIM, audit logs, private endpoints and Bring Your Own Cloud on Enterprise.

Where it falls short

  • Light use pays the floor: Standard carries a US$50 monthly minimum, and one Reddit user with a few dollars of yearly usage was told a US$50 monthly minimum spend would apply to the account (July 2025).
  • SDK, CLI and Terraform defects draw the most public reports, from silent failures on IDs with spaces in Python SDK v10 to a Terraform provider that never converges on integrated sparse indexes.
  • It runs as a hosted service: one PeerSpot reviewer reports no on-premises option for highly regulated industries, and a Head of Engineering passed on Pinecone mainly because its serverless region was only in the United States.
  • Operation limits apply per request, and Omi's account deletions failed in production when one call exceeded 1,000 ids (August 2026).

Implementation reality

  1. Price the workload

    Standard lists storage at US$0.33 per GB a month, write units at US$4 to US$4.50 per million and read units at US$16 to US$18 per million, varying by cloud and region. A query uses 1 read unit for every 1 GB of namespace size, with a minimum of 0.25. Usage past the US$50 minimum is billed pay-as-you-go.

  2. Pick plan and region

    Starter indexes run in AWS us-east-1 only, and a serverless index's cloud and region cannot change after creation. Builder and above can use any supported region on AWS, GCP or Azure, and Bring Your Own Cloud on Enterprise keeps vectors and payloads in your own account.

  3. Pin SDKs and respect limits

    Pin SDK and Terraform provider versions and test upgrades: public reports cover defects and gaps across the Python, Java, TypeScript, .NET and Go clients. An upsert batch holds 1,000 records and a delete takes 1,000 ids, so chunk bulk operations.

  4. Set identity and logs

    SSO is available on Standard and Enterprise, SCIM on Enterprise, and audit logs on Enterprise or with the HIPAA add-on at US$190 a month on Standard with a 6-month minimum.

What to verify before you buy

Plans and billing

  • On Standard, the US$50 monthly minimum applies to usage, and anything over US$50 is billed pay-as-you-go. On Enterprise, once usage exceeds the US$500 minimum, you pay as you go.
  • Standard and Enterprise include 100 GB of egress a month, and egress beyond the allowance is billed at the per-GB overage rate of US$0.10 per GB while reads keep serving. On Starter and Builder, in-scope reads are blocked with a 429 error once the allowance is reached.
  • Customers can buy between US$8,000 and US$25,000 in prepaid credits, which apply at List Price, and usage beyond the available prepaid credits is billed at full List Price.
  • The Builder plan is a flat US$20 a month plan with higher quotas than Starter and no usage overages.
  • On Standard and Enterprise, llama-text-embed-v2 costs US$0.16 per million tokens, multilingual-e5-large and pinecone-sparse-english-v0 US$0.08 per million tokens, and bge-reranker-v2-m3 and pinecone-rerank-v0 US$2 per 1,000 requests. Starter includes 5 million and Builder 10 million embedding tokens a month per model.
  • On Standard and Enterprise, storage costs US$0.33 per GB a month, import from object storage US$0.25 per GB, backups US$0.10 per GB a month and restore from backup US$0.15 per GB.
  • A BYOC bill is a flat monthly platform fee for each environment plus a rate for each node, metered in node hours, and you pay your cloud provider directly for the underlying infrastructure. Current rates come from Pinecone on request.
  • Pinecone is available on the AWS, GCP and Microsoft cloud marketplaces, and an organization can set up pay-as-you-go billing through AWS Marketplace. Prepaid credits can't be bought through marketplace billing.

Identity, security and data

  • SSO is available on Standard and Enterprise plans, and the Okta setup can be adapted for any provider with SAML 2.0 support. You can choose to enforce SSO, so all members of your organization must use it to log in.
  • SCIM provisioning is available on the Enterprise plan and builds on SAML SSO. Changes made in Okta sync to Pinecone over SCIM in near real time, and SCIM group push isn't supported.
  • Audit logs record user, service account and API actions on the management and control plane and are written to an Amazon S3 bucket every 30 minutes. They are available only on Enterprise plans or with the HIPAA compliance add-on, and only an organization owner can enable them.
  • Pinecone captures audit log events within two hours of occurrence and retains them for 90 days, after which they are permanently deleted.
  • Pinecone is SOC2 Type II certified.
  • Pinecone is HIPAA compliant, with a BAA available upon request. Full HIPAA compliance is included with Enterprise, Standard offers it as a US$190 per month add-on with a 6-month minimum, and on either it requires configuring audit logs.
  • Pinecone's Trust Center lists ISO 27001:2022 under Compliance and offers a Pinecone ISO 27001 2025 Certificate among its resources.
  • After a deletion request, Pinecone marks the data for deletion and retains it for a maximum of 90 days, inaccessible to any user, and permanently deletes it before the end of that window.
  • If an account is deactivated for non-payment, account data is retained for 30 days and then permanently deleted. After termination, on the customer's written request, an End User may access the Services for up to 30 days to retrieve Customer Data, and then, unless prohibited by Applicable Law, Pinecone deletes Customer Data in accordance with the Documentation.
  • Under the Master Subscription Agreement, the customer grants Pinecone the right to use and process Customer Data solely to provide Services and as otherwise expressly provided in the agreement. The security page says data is only used for servicing API calls.
  • Pinecone encrypts data at rest and in transit with AES-256. On the Enterprise plan, private endpoints reach indexes through AWS PrivateLink or Azure Private Link and carry data plane traffic only, and customer-managed encryption keys use AWS KMS.

Limits

  • Operation limits are fixed and don't vary by plan: an upsert batch holds 1,000 records with vectors, up to 2 MB in total, or 96 records with text, a query takes a top_k of at most 10,000 and returns at most 4MB, and fetch and delete requests take at most 1,000 record IDs.
  • Indexes are limited to 5 on Starter and 10, 20 and 200 per project on Builder, Standard and Enterprise. Namespaces per index are 100, 1,000, 100,000 and 1,000,000. Exceeding an object limit returns 403 QUOTA_EXCEEDED, and most object limits can be raised on request.
  • On serverless indexes, Starter allows 1,000,000 read units and 2,000,000 write units a month per organization, and Builder 2,000,000 and 5,000,000. Standard and Enterprise are unlimited. Reaching a monthly usage limit returns a 429 error.
  • Rate limits apply to serverless indexes only. On every plan, queries are limited to 100 requests per second per namespace and 2,000 read units per second per index, and upserts to 100 requests and 50 MB per second per namespace. Most rate limits can be adjusted on request, and indexes on Dedicated Read Nodes aren't subject to read unit limits for query, fetch, list and full-text search.
  • llama-text-embed-v2 and multilingual-e5-large embed up to 250,000 passage tokens a minute on Starter and Builder and 1,000,000 on Standard and Enterprise. Inference requests are limited to 100 per second and 2,000 per minute per project on every plan.
  • Pinecone is eventually consistent, so there can be a slight delay before upserted records are available to query.

Deployment

  • Starter serverless indexes must be in the AWS us-east-1 region. Builder, Standard and Enterprise can create indexes in any supported region, on AWS, Azure or GCP.
  • The models table lists the processing location of llama-text-embed-v2, multilingual-e5-large, pinecone-sparse-english-v0 and bge-reranker-v2-m3 as us, and of cohere-rerank-4-fast as global.
  • In BYOC, the console index browser can't read or write data, the Pinecone CLI can't run data plane operations, and imports from private buckets need the bucket in the same cloud account.
  • In BYOC, vectors, metadata, index contents and query and upsert payloads do not leave your cloud account. Only operational metrics and traces are sent to Pinecone, with customer data filtered out before transmission.

What to measure in your pilot

  • The Master Subscription Agreement lets Pinecone use and process Customer Data solely to provide the Services and as the agreement provides; confirm training terms in your own contract.
  • Bring Your Own Cloud rates come from Pinecone on request; price them before you plan an Enterprise deployment.
  • Public latency figures here are single accounts on small corpora; measure p99 on your own data and region.

Sources

Official documentation
33
Public source
3
  1. Pinecone: Understanding costOfficial documentationdocs.pinecone.io
  2. Pinecone: Standard trialOfficial documentationdocs.pinecone.io
  3. Pinecone: Upgrade billing planOfficial documentationdocs.pinecone.io
  4. Pinecone: OverviewOfficial documentationdocs.pinecone.io
  5. Pinecone: PricingOfficial documentationpinecone.io
  6. Pinecone: Bring Your Own Cloud: deployOfficial documentationdocs.pinecone.io
  7. Pinecone: Index configurationOfficial documentationdocs.pinecone.io
  8. Pinecone: ModelsOfficial documentationdocs.pinecone.io
  9. Pinecone: Integrated embeddingOfficial documentationdocs.pinecone.io
  10. Pinecone: Bring your own vectorsOfficial documentationdocs.pinecone.io
  11. Pinecone: Operation limitsOfficial documentationdocs.pinecone.io
  12. Pinecone: AWS MarketplaceOfficial documentationdocs.pinecone.io
  13. Pinecone: LangChainOfficial documentationdocs.pinecone.io
  14. Pinecone: LlamaIndexOfficial documentationdocs.pinecone.io
  15. Pinecone: MCP serverOfficial documentationdocs.pinecone.io
  16. Pinecone: Amazon BedrockOfficial documentationdocs.pinecone.io
  17. Pinecone: Bring Your Own CloudOfficial documentationdocs.pinecone.io
  18. Pinecone: Object limitsOfficial documentationdocs.pinecone.io
  19. Pinecone: SecurityOfficial documentationpinecone.io
  20. Pinecone: Configure private endpointsOfficial documentationdocs.pinecone.io
  21. Pinecone: Customer-managed encryption keysOfficial documentationdocs.pinecone.io
  22. Pinecone: Security overviewOfficial documentationdocs.pinecone.io
  23. Pinecone: Data deletionOfficial documentationdocs.pinecone.io
  24. Pinecone: SSO with OktaOfficial documentationdocs.pinecone.io
  25. Pinecone: SCIM with OktaOfficial documentationdocs.pinecone.io
  26. Pinecone: Configure audit logsOfficial documentationdocs.pinecone.io
  27. Pinecone: Trust CenterOfficial documentationsecurity.pinecone.io
  28. Pinecone: Non-payment account deactivationOfficial documentationdocs.pinecone.io
  29. Pinecone: Master Subscription AgreementOfficial documentationpinecone.io
  30. Pinecone: Rate limitsOfficial documentationdocs.pinecone.io
  31. Pinecone: Known limitationsOfficial documentationdocs.pinecone.io
  32. Pinecone: ArchitectureOfficial documentationdocs.pinecone.io
  33. Pinecone: llms-full.txtOfficial documentationdocs.pinecone.io
  34. Reddit, realam1: a US$50 monthly minimum on light use (Jul 17, 2025)Public sourcereddit.com
  35. BasedHardware/omi pull request #12466: chunk Pinecone deletes (Aug 31, 2026)Public sourcegithub.com
  36. pinecone-io/python-sdk issue #431: ARM images for Pinecone Local (Jan 17, 2025)Public sourcegithub.com