Reranking model reference link changed from 'Reranking models' to 'cohere-rerank-4-fast'
Users migrating rerank requests should reference the updated documentation link that directly points to the cohere-rerank-4-fast model page instead of a general reranking models page.
Evidence
−For model details and parameters, see Reranking models.
+For model details and parameters, see cohere-rerank-4-fast.
Doubled included embedding tokens through December 31, 2026 promotional offer
Starter and Builder plan customers get 2x the embedded tokens at no additional cost through end of 2026: Starter 10M tokens/month/model (up from 5M), Builder 20M tokens/month/model (up from 10M).
Evidence
+Twice the included embedding tokens through December 31
From November 4 through December 31, 2026, Starter and Builder organizations get twice the included tokens on every embedding model. Starter includes 10,000,000 tokens per month per model (5,000,000 before the promotion), and Builder includes 20,000,000 (10,000,000 before).
New embedding models for document indexes: Google Gemini and OpenAI text-embedding-3
Users can now choose from three additional hosted embedding models (gemini-embedding-2, text-embedding-3-small, text-embedding-3-large) for document indexes, with support for up to 3,072 dimensions and 8,192 tokens per value. Gemini model available on Builder/Standard/Enterprise plans; text-embedding-3 models available on all plans.
Evidence
+Document indexes can now embed text with three more hosted models: gemini-embedding-2 from Google, and text-embedding-3-small and text-embedding-3-large from OpenAI. Each takes any dimension in a range, up to 3,072, and embeds up to 8,192 tokens per value.
Terraform Provider v5.0.0 major version removes pod-based index creation support
Users cannot create new pod-based indexes with v5.0.0. Requires migration steps for affected configurations. Breaking change requiring upgrade planning.
Evidence
+This is a major version. New pod-based indexes can't be created, and a few configurations that earlier versions accepted are now rejected at plan time.
New cross-region transfer fee introduced for database backup restores.
Users restoring backups to different regions will incur additional per-GB charges based on cloud provider and backup region. Same-region restores remain unchanged.
Evidence
+Cross-region transfer fee for restores
Restoring a backup into a different region now incurs a cross-region transfer fee on top of the standard restore price. The fee is charged per GB and depends on the cloud provider and the backup's region. Restoring into the same region as the backup is billed as before.The fee appears on your invoice as Serverless Cross-Region Transfer.
Added guidance to mock the Pinecone API for testing without an account
Developers seeking alternatives to Pinecone Local now have an explicit recommendation to use API mocking instead of requiring a separate project account.
Evidence
−Pinecone Local, the in-memory Docker emulator for local development and testing, has reached end of life and is no longer available. Its documentation has been removed.To develop and test your applications, use a separate project in your Pinecone account.
+Pinecone Local, the in-memory Docker emulator for local development and testing, has reached end of life and is no longer available. Its documentation has been removed.To develop and test your applications, use a separate project in your Pinecone account. To test your client code without an account, mock the Pinecone API instead.
Pinecone Local emulator reaches end of life and is no longer available
Developers who use Pinecone Local for local development and testing must migrate to using a separate Pinecone project instead. This affects the developer workflow for local development and automated testing.
Evidence
−Pinecone Local (local development emulator) — Local development
Automated testing with Pinecone Local — Automated testing
+End of life: Pinecone Local
Pinecone Local, the in-memory Docker emulator for local development and testing, has reached end of life and is no longer available. Its documentation has been removed.To develop and test your applications, use a separate project in your Pinecone account.
Pinecone Go SDK v7.0.0 released with schema-based document indexes and Documents API support
Go SDK users must upgrade and migrate code due to breaking changes including module path change, removal of pod-based index creation, enum constant prefixing, and API parameter changes.
Evidence
+Released v7.0.0 of the Pinecone Go SDK. This version uses the latest stable API version, 2026-07, and adds support for schema-based document indexes and the Documents API, which brings full-text search to the Go SDK.
Namespace aliases feature released for Pinecone database
Users can now create stable, repointable aliases for namespaces that allow atomic data swaps without application redeploys, improving deployment flexibility and reducing downtime during namespace migrations.
Evidence
+Namespace aliases
A namespace alias is a stable, repointable name that points at a namespace. Your application reads through the alias, and you can swap the data it reads by atomically repointing the alias to a different namespace in the same index, with no redeploy. Manage aliases through the REST API on version 2026-07 or later.
Pinecone Bring Your Own Cloud now Generally Available (replaces Full-Text Search announcement)
Pinecone is promoting a new cloud deployment capability on their pricing page. Users now see information about BYOC availability instead of the previous Full-Text Search feature announcement.
Evidence
−Full-Text Search is Now Generally Available In Pinecone Database - Read the announcementDismiss
+Pinecone Bring Your Own Cloud is Now Generally Available - Read the announcementDismiss
Changed HIPAA add-on enablement from contacting sales to submitting a HIPAA request
Standard plan customers now have a self-service option (HIPAA request) in addition to or instead of contacting sales to enable HIPAA compliance add-on.
Evidence
−To enable the add-on on the Standard plan, contact sales or see Understanding cost — HIPAA compliance add-on.
+To enable the add-on on the Standard plan, submit a HIPAA request or see Understanding cost — HIPAA compliance add-on.
Gemini 2.5 Pro and o4-mini models deprecated; automatic routing to newer models implemented
Existing code using gemini-2.5-pro will automatically route to Gemini 3.5 Flash, and o4-mini requests will route to GPT-5 at the same price. Users are encouraged to update code explicitly but no changes are required.
Evidence
+Gemini 2.5 Pro and o4-mini deprecations for Assistant
Google has deprecated the Gemini 2.5 Pro model, and OpenAI has deprecated o4-mini. Pinecone Assistant automatically routes all chat requests that specify gemini-2.5-pro to Gemini 3.5 Flash, and all chat requests that specify o4-mini to GPT-5, at the same price.
Enterprise plan now supports 1,000,000 namespaces per serverless index
Enterprise customers can now use significantly more namespaces in their serverless indexes, with the ability to request even higher limits for specific use cases.
Evidence
+Higher namespace limit on the Enterprise plan
The Enterprise limit for namespaces per serverless index is now 1,000,000. Pinecone can accommodate more for specific use cases. Contact Support to request a higher limit.
Added comparison messaging to pricing page highlighting Pinecone against pgvector, Elasticsearch, Vertex AI and other alternatives.
Users evaluating Pinecone pricing now see direct comparisons with competing solutions, potentially strengthening positioning against alternatives like pgvector, Elasticsearch, and Vertex AI.
Evidence
+Weighing this against what you run today? The Pinecone comparisons cover pgvector, Elasticsearch, Vertex AI, and more.
Full-text search and Documents API reach general availability
Developers can now use production-ready full-text search (BM25) combined with dense and sparse vector search in single indexes, with advanced querying, filtering, and bulk document operations.
Evidence
+General availability: Full-text search and the Documents API
Full-text search is now generally available on API version 2026-07, built on the new Documents API.Create schema-based document indexes that combine full-text (BM25), dense-vector, and sparse-vector ranking fields in a single index.
Bulk import credit of $250 (1 TB) no longer offered for new subscriptions
New Standard and Enterprise customers will no longer receive the one-time $250 bulk import credit and will be billed at $0.25/GB for all imports from object storage.
Evidence
+Bulk import credit is no longer offered
New Standard and Enterprise subscriptions no longer receive the one-time $250 bulk import credit (1 TB). Imports from object storage are billed at the standard $0.25/GB rate.
cohere-rerank-3.5 deprecation deadline moved from July 31 to August 31, 2026
Customers have one additional month (extended from July 31 to August 31) to migrate rerank requests to cohere-rerank-4-fast before automatic migration occurs.
Evidence
−Through July 31, 2026: cohere-rerank-3.5 continues to serve requests as before.
August 1, 2026: requests to cohere-rerank-3.5 are automatically served by cohere-rerank-4-fast.
+Through August 30, 2026: cohere-rerank-3.5 continues to serve requests as before.
August 31, 2026: requests to cohere-rerank-3.5 are automatically served by cohere-rerank-4-fast.
Cohere reranking API updated from v3.5 to 4-fast with pricing unit terminology change
Developers using Cohere reranking through Pinecone will need to update integrations to use the new cohere-rerank-4-fast model. Pricing unit terminology changed from 'requests' to 'rerank units' at the same $2/1k rate.
Index fullness metrics now available for dedicated read nodes via Prometheus/Datadog integration
Users with dedicated read nodes can now monitor index fullness over time through monitoring integrations, enabling better capacity planning.
Evidence
+Index fullness metrics for dedicated read nodes
Indexes that use dedicated read nodes now emit index fullness as time-series metrics. Monitor fullness over time via the Prometheus or Datadog integration.
New reciprocal rank fusion (RRF) guide removed from release notes
The RRF guide announcement is no longer featured in the 2026 release notes.
Evidence
−2026-08-11
Search
New guide: reciprocal rank fusion (RRF)
Added a guide on reciprocal rank fusion, a client-side method for combining the results of separate searches (for example, full-text and semantic) into one ranking without normalizing scores. RRF fuses rankings rather than raw scores, so it needs no per-signal weighting.
New guide published on reciprocal rank fusion (RRF) for search result ranking
Developers can now learn how to combine results from multiple search methods (full-text and semantic) using RRF, a client-side ranking technique that fuses rankings without requiring score normalization or per-signal weighting.
Evidence
+New guide: reciprocal rank fusion (RRF)
Added a guide on reciprocal rank fusion, a client-side method for combining the results of separate searches (for example, full-text and semantic) into one ranking without normalizing scores. RRF fuses rankings rather than raw scores, so it needs no per-signal weighting.
Pinecone Terraform Provider v4.0.0 released with service account and organization management resources
Users can now manage service accounts, role bindings, organization invites, and members through Terraform, with new data sources and import support for all resources.
Evidence
+Released Pinecone Terraform Provider v4.0.0
Released v4.0.0 of the Terraform Provider for Pinecone. This version adds resources for managing service accounts, role bindings, organization invites, and removing organization members. It also adds data sources for reading existing indexes, collections, projects, service accounts, role bindings, invites, and users, and terraform import support for every resource.
Full-text search now supports fuzzy matching with typo tolerance
Users can now perform typo-tolerant searches in query_string searches by appending ~ to terms, with configurable edit distance (0-2) or automatic distance based on term length. No reindexing required on existing indexes.
Evidence
+Fuzzy matching (typo tolerance) — In a query_string search, append ~ to a term to match indexed terms within a small edit distance: body:(compxter~1) matches computer. Use term~ for an automatic distance based on term length, or term~N for a fixed distance (N is 0–2). Fuzzy works on existing indexes with no reindex.
Role naming change: 'User' role renamed to 'Manager' for organizations and projects
Existing role assignments are automatically migrated. The organization-level User role is now Organization manager, and project-level User role is now Project manager. Permissions remain unchanged.
Evidence
+If you used the previous role model, the User role has been renamed to Manager: the organization-level User role is now Organization manager, and the project-level User role is now Project manager. This is a name change only—permissions are unchanged and existing assignments are migrated automatically.
Bring Your Own Cloud (BYOC) option added to Pinecone pricing page
Users now have the option to deploy Pinecone on their own cloud infrastructure, providing more control and flexibility over data residency and compliance requirements.
Evidence
+Bring Your Own Cloud (BYOC)
Bring Your Own Cloud (BYOC)