Managed services - verified
turbopuffer
turbopuffer is listed under managed services (deployment type: managed serverless). First listed price, in its own unit: $16, minimum usage per month (launch plan). Per-GB storage, write and query rates appear only in a pricing calculator. Max dimensions: 10,752 (dense vectors). Stated: hybrid and sparse search and built-in embedding. Read from the vendor's own pages on October 18, 2026.
Read from turbopuffer.com. Docs - Pricing - Compare every vector database
Status
- Availability label
- Not stated
No GA or preview label for the service is stated on the pricing, limits, query or write pages checked (the pricing page carries a banner that turbopuffer v3 is coming).
How turbopuffer charges
Every price turbopuffer states, in its own unit, with the quote and where it comes from. Nothing is converted.
| Price | Unit, as stated | Quote and source | Where it comes from |
|---|---|---|---|
| Plan price or monthly minimum | |||
| $16 | minimum usage per month (launch plan) | “Sign up $16/month minimum usage scale” | - |
| $256 | minimum usage per month (scale plan) | “Sign up $256/month minimum usage enterprise” | - |
| $4,096 | per month, with a 35% usage premium (enterprise plan) | “Contact us =$4,096/month 35% usage premium” | - |
| Product | Rate | Why | Billing model, quoted |
|---|---|---|---|
| turbopuffer | Not statedPer-GB storage, write and query rates | Per-GB storage, write and query rates appear only in the pricing page's cost calculator, not in the page text, so no per-unit rate is listed here. | See the plan minimums above. |
Per-GB storage, write and query rates appear only in the pricing page's cost calculator, not in the page text, so no per-unit rate is listed here.
Plan labels in brackets come from the pricing table's column headings, which the quoted cell text does not always repeat.
Free tier and limits
- Free tier
- Not stated
No free plan is stated on the pricing page; the lowest plan carries a $16/month minimum.
- Limits
Up to 128B documents (256 TB) per namespace
“Max documents (per namespace) 128B @ 256TB( seen: 100B @ 200TB)”
turbopuffer.com
Search and indexing
- Max dimensions
10,752
Applies to: dense vectors
“Max dimensions for dense vectors 10,752”
turbopuffer.com - Distance metrics
cosine_distance, euclidean_squared
“Possible values are cosine_distance or euclidean_squared.”
turbopuffer.com - Index types
- Not stated
The query docs describe ANN, exact kNN, BM25 and sparse search but do not name the ANN index algorithm.
- Hybrid and sparse search
Yes: BM25 full-text search and sparse vector search alongside ANN
“Full-text search: find documents with the highest BM25 score, a classic text search algorithm that considers query term frequency and document length”
turbopuffer.com “Supported ranking functions: - ANN ("approximate nearest neighbor") - kNN ("exact nearest neighbor", requires filters) - BM25 (combine with Sum, Max) - SparseKNN (sparse vector search)”
turbopuffer.com - Metadata filtering
Yes
“A query retrieves documents in a single namespace, returning the ordered or highest-ranked documents that match the query's filters.”
turbopuffer.com - Multi-tenancy
Namespaces; count listed as unlimited on the limits page
“Max number of namespaces Unlimited (seen: 250M+)”
turbopuffer.com
Built-in embedding and reranking
- Built-in embedding
Native embedding models can be enabled per attribute in the schema
“embedding_tokens (uint): the number of tokens embedded, only set when using a native embedding model”
turbopuffer.com “configure full-text search for an attribute, disable filtering for an attribute, or enable native embedding.”
turbopuffer.com - Built-in reranking
- Not stated
Not stated on the query, write or pricing pages checked.
Deployment, license and SDKs
- Regions and clouds
AWS, GCP and Azure (azure-eastus2 in private preview); see the regions page for the full list
“aws-us-east-1 N. Virginia, US”
turbopuffer.com “gcp-us-central1 Iowa, US”
turbopuffer.com “azure-eastus2 Private preview, contact us for access Virginia, US”
turbopuffer.com - BYOC and self-host
BYOC in AWS, GCP and Azure; single-tenancy clusters
“In addition to these public clusters, we run dedicated clusters in various other regions for single-tenancy customers and in any region inside your VPC in AWS, GCP and Azure (BYOC).”
turbopuffer.com - License
- Not stated
No open-source license is stated on the pages checked.
- SDKs
- Not stated
An official SDK list is not stated on the pages checked (the regions page shows code tabs, which are not treated as an SDK list here).
Vendor-stated claims
Performance, scale and SLA figures as turbopuffer states them. Not measured.
Vector search recall@10 of 90-100% (limits page)vendor-stated
“Vector search recall@10 90-100%”
5k+ queries/s per namespacevendor-stated
“Max queries (per namespace) 5k+ queries/s”
99.95% uptime SLA on enterprisevendor-statedinferred
Our reading: The quote does not name the plan; enterprise comes from the pricing table column it sits in.
“Uptime SLA99.95%”
Research notes
Plans differ by minimum monthly usage; the pricing page says all database features are included in every plan. The pricing page banner reads "turbopuffer v3 is coming".
FAQ
How much does turbopuffer cost?
In turbopuffer's own units: $16 (minimum usage per month (launch plan)); $256 (minimum usage per month (scale plan)); $4,096 (per month, with a 35% usage premium (enterprise plan)). Per-GB storage, write and query rates: not stated. Per-GB storage, write and query rates appear only in the pricing page's cost calculator, not in the page text, so no per-unit rate is listed here. As of October 18, 2026; prices are not converted.
What is the maximum vector dimension in turbopuffer?
10,752 (dense vectors), per the vendor: "Max dimensions for dense vectors 10,752"
Does turbopuffer support hybrid search?
Yes: BM25 full-text search and sparse vector search alongside ANN, per the vendor: "Full-text search: find documents with the highest BM25 score, a classic text search algorithm that considers query term frequency and document length"
Does turbopuffer have built-in embedding or reranking?
Embedding: Native embedding models can be enabled per attribute in the schema. Reranking: not stated. Not stated on the query, write or pricing pages checked.
Sources
- Source: turbopuffer.com/docs
- Source: turbopuffer.com/pricing
- Source: turbopuffer.com/docs/limits
- Source: turbopuffer.com/docs/write
- Source: turbopuffer.com/docs/query
- Source: turbopuffer.com/docs/regions
Last verified from each vendor's own docs, pricing pages, READMEs and license files; every quote was string-matched against the page that day. This is a dated snapshot, not a benchmark: prices are shown in each product's own unit and change without notice, so check the linked source before you commit.
Newsletter
Know when a vector database changes its pricing
One email per week. New model releases, retrieval tooling and pricing changes and what they mean for what you build.