kitio.aiAdded Oct 28, 2026
Turbopuffer
From $16/moServerless vector and full-text search built on object storage
What is Turbopuffer?
Turbopuffer is a serverless search engine for vectors and full text that stores data on object storage with a cache in front. This design makes it much cheaper for very large, multi-tenant datasets, and it is used by several well-known AI apps.
Namespaces scale to huge counts, which suits apps that keep a separate index per user or workspace. It supports hybrid search, filtering and fast queries on warm data, with slower first queries on cold data.
Pricing is usage-based for storage, writes and queries, with a monthly minimum on paid plans. There is no open-source version, so it cannot be self-hosted on your own servers.
Features
- Object storage backend
- Vector and full-text search
- Massive namespace counts
- Metadata filtering
- Usage-based pricing
Tags
Releases
+ Submit new release- Listed
Turbopuffer was added to the directory.
Organization
Turbopuffer
Listed by kitio.ai · unverified owner
Pricing
- Pricing model
- Paid
- Paid options from
- $16/mo
- Billing frequency
- Monthly
Turbopuffer plans
Launch
$16/mo
- Minimum usage, billed on usage
- All database features
- Community Slack and email support
Scale
$256/mo
- Minimum usage
- HIPAA-ready BAA
- SSO and audit logs
- Private Slack channel
Enterprise
From $4,096/mo
- Single-tenancy and BYOC
- Private networking
- 24/7 support with SLA
Prices in USD, checked Oct 28, 2026. Plans change often, so confirm on the Turbopuffer site.Official pricing
Sponsored
Your AI tool here
A Featured launch puts your tool in this spot and on the home page for a week.
Top Turbopuffer alternatives
View allModels & APIs
Managed serverless vector database for search and RAG
Models & APIs
Serverless vector database with per-request pricing and a free tier
Models & APIs
Serverless inference for open models on Cloudflare's global network
Compare Turbopuffer
If you liked Turbopuffer
On-demand GPU pods and serverless endpoints for AI workloads
PaidModels & APIsMade in United StatesServerless cloud for running Python and GPU workloads on demand
Freemium · $250/moModels & APIsMade in United StatesOpen-source vector database with hybrid search and built-in vectorizers
Freemium · $45/moModels & APIsMade in NetherlandsOpen-source vector database built for billion-scale search
FreemiumModels & APIsMade in United States
Pros and cons
Pros
- Very low storage cost
- Great for per-user indexes
- Simple API
Cons
- Cold queries are slower
- Monthly minimum spend
- Closed source
Prompts & results
No prompts yet. Add your own prompt and output to help others use this tool.
Share a prompt
Reviews
Q&A
What is Turbopuffer?
Turbopuffer is a serverless search engine for vectors and full text that stores data on object storage with a cache in front. This design makes it much cheaper for very large, multi-tenant datasets, and it is used by several well-known AI apps.
How much does Turbopuffer cost?
Turbopuffer is paid, with paid plans from $16/mo.
Is there a free version of Turbopuffer?
No. Turbopuffer is a paid tool.
What are the main features of Turbopuffer?
Object storage backend, Vector and full-text search, Massive namespace counts, Metadata filtering, Usage-based pricing.
Who makes Turbopuffer?
Turbopuffer is made by Turbopuffer.
What are the best alternatives to Turbopuffer?
Popular alternatives include Pinecone, Upstash Vector, SvectorDB.