Turbopuffer vs Qdrant
SEPT 2026 auditA comparison of Turbopuffer and Qdrant built from values read directly from each provider's own documentation, with the source recorded against every figure.
The short answer
Choose Turbopuffer if…
Teams that want vector search billed like storage rather than compute — object-storage-backed pricing tends to undercut memory-resident vector databases at rest, especially for large, cold datasets.
Editorial · Palash Bagchi · approved
Choose Qdrant if…
Teams that want a choice of similarity metric rather than one fixed default, or that want the option to self-host the same open-source engine the managed cloud runs.
Editorial · Palash Bagchi · approved
Consider something else if…
- Turbopuffer: The full list of supported similarity metrics matters before committing — that wasn't fully documented in what's sourced here, unlike competitors that publish the complete list.
- Qdrant: A published, calculator-free entry price matters — Qdrant Cloud's Standard tier pricing is usage-based with no static per-month figure quoted, unlike some competitors' flat entry plans.
Neither quite fits? Every other option, ranked by iScore: Turbopuffer alternatives · Qdrant alternatives
18 sourced criteria separate them — see where, with sources, below.
Pick the criteria you care about. The chart counts how many of them lean toward each provider — the same read as scanning the bars below, just totalled for the ones you chose.
Turbopuffer 0
Qdrant 0
At a glance.
| Criterion | Turbopuffer | Qdrant |
|---|---|---|
| Pricing | ||
| Free tier | No — Entry plan (launch) carries a $16/month minimum usage; pricing lists no free plan. | Yes |
| Kind of free offer | No free offer | Always free |
| Entry paid plan | Launch plan: $16/month minimum usage, including all database features. Scale plan: $256/month minimum. Enterprise: $4,096/month or more, plus a 35% usage premium. | Smallest paid Managed Cloud cluster (1 node, 0.5 vCPU, 2 GiB RAM, 8 GiB disk) on AWS us-east-1: $0.03566/hour, about $26.10/month, per Qdrant's own Cloud pricing calculator. Calculator estimate for AWS us-east-1; VAT excluded and expressly not a binding offer. |
| Pricing model | Usage-based, against a monthly minimum per plan: $16 Launch, $256 Scale, $4,096 or more for Enterprise, which the page also marks with a 35% usage premium. | Usage-based on the Standard tier: billed for compute (vCPU), memory (GB) and storage (GB) consumed by clusters, backup storage, and inference tokens for paid models. |
| Infrastructure | ||
| Regions | 20 public regions listed in the docs: 10 on AWS, 9 on GCP and 1 on Azure (eastus2); dedicated and BYOC clusters elsewhere on request. | Standard and Premium tiers run on AWS, Azure and GCP. |
| Uptime SLA | 99.95%, listed only against Enterprise (from $4,096/month). Launch and Scale publish no uptime SLA. | 99.5% on the paid cloud tier; 99.9% on Premium. |
| Vector | ||
| Max vector dimensions | 10,752 maximum dimensions for dense vectors; 1,024 maximum dimensions per sparse vector (sparse total dimensions unlimited) | Up to 65,535 dimensions for dense vectors |
| Similarity metrics supported | Cosine distance is documented as a supported metric (used in the SDK's example code); the full list of supported metrics wasn't confirmed beyond this. | Dot product, Cosine similarity (implemented as dot-product over automatically normalized vectors), Euclidean distance and Manhattan distance. |
| Hybrid search (vector + keyword) | Supported | Native hybrid queries combine dense semantic and sparse lexical (BM25) vectors in a single query |
| Metadata filtering | Exact attribute filters on every query (and/or/not, eq, in, contains, range); the planner combines the attribute index with the ANN index. | Supported: conditions can be set on the payload and on the point id, combined with AND, OR and NOT clauses that nest arbitrarily. |
| Compliance | ||
| SOC 2 | SOC 2 Type 2 audited for security and availability controls; report and penetration test available on request from the Trust Center. | SOC 2 Type 2 certified; compliance reports available on request through the Qdrant Trust Center. |
| HIPAA | Customers storing PHI may request a Business Associate Agreement; turbopuffer commits to the HIPAA requirements on business associates. Pricing page lists the HIPAA-ready BAA on the Scale and Enterprise plans, not on Launch. | HIPAA certified; a Business Associate Agreement is available for Qdrant Managed Cloud. |
| GDPR / data residency | Data stays in the region selected unless the customer moves it via the API; GDPR/CCPA Data Processing Addendum offered to all customers. | DPA offered for GDPR; cluster data is stored only in the cluster's deployment region. |
| Encryption at rest | Customer data always encrypted at rest with AES-256, and in transit with TLS 1.2+; optionally with a customer's own key. | All storage volumes in Qdrant Managed Cloud are encrypted at rest; data in transit is protected with TLS. |
| Customer-managed keys | CMEK on the Enterprise plan using keys from the customer's own KMS/EKM; the encryption key is defined at the namespace level. | Premium tier customers can bring their own encryption keys for storage volumes. |
| Role-based access control | No built-in row or document-level RBAC; docs say permissions have to be implemented at the user level with query filters. Statement is about row/document-level RBAC; no org-level role model is documented either way. | Built-in RBAC in the Qdrant Cloud Management Console, with permissions manageable at collection level. |
| Private networking / VPC | Private connections from customer VPCs to multi-tenant regions on the Enterprise plan: AWS PrivateLink, GCP Private Service Connect. | Private link connectivity from a Managed Cloud cluster to your VPC, available to Premium tier customers only. |
| Dedicated infrastructure | Dedicated single-tenancy clusters in further regions, or BYOC clusters inside the customer's own VPC on AWS, GCP or Azure, spun up on request. Pricing page lists Single-Tenancy and BYOC under Deployment on the Enterprise plan only. | Paid Managed Cloud clusters run on their own dedicated resources, isolated in hardened, unprivileged containers. |
Where they differ.
Free tier
- Turbopuffer
- No — Entry plan (launch) carries a $16/month minimum usage; pricing lists no free plan.
- Qdrant
- Yes
Sources (2) →Sources ↓
- turbopuffer pricing ↗
“turbopuffer pricing launch Includes all database features Deployment Multi-Tenancy Security & Compliance SOC2 report, GDPR-ready DPA Support Community Slack & Email Sign up $16/month minimum usage”
Read 2026-09-13 · official pricing · Report a correction
- Qdrant Pricing ↗
“The free tier includes 1GB RAM and 4GB disk storage, upgrading will get you a dedicated cluster.”
Read 2026-09-13 · official pricing · Report a correction
Kind of free offer
- Turbopuffer
- No free offer
- Qdrant
- Always free
Sources (2) →Sources ↓
- turbopuffer pricing ↗
“turbopuffer pricing launch Includes all database features Deployment Multi-Tenancy Security & Com”
Read 2026-09-13 · official pricing · Report a correction
- Qdrant Pricing ↗
“Free Tier Free forever For testing, and prototypes Single Node Cluster 0.5 vCPU / 1GB RAM/ 4 GB Disk. Free Cloud Inference With Selected Models Start Free”
Read 2026-09-14 · official pricing · Report a correction
Entry paid plan
- Turbopuffer
- Launch plan: $16/month minimum usage, including all database features. Scale plan: $256/month minimum. Enterprise: $4,096/month or more, plus a 35% usage premium.
- Qdrant
- Smallest paid Managed Cloud cluster (1 node, 0.5 vCPU, 2 GiB RAM, 8 GiB disk) on AWS us-east-1: $0.03566/hour, about $26.10/month, per Qdrant's own Cloud pricing calculator. Calculator estimate for AWS us-east-1; VAT excluded and expressly not a binding offer.
Sources (2) →Sources ↓
- Turbopuffer Pricing ↗
“$16/month minimum usage [...] $256/month minimum usage [...] Contact us >= $4,096/month 35% usage premium”
Read 2026-09-13 · official pricing · Report a correction
- Qdrant Cloud · Calculator ↗
“RAM: 2 GiB vCPUs: 0.5 vCPU Disk Space: 8 GiB Nodes: 1 Node [...] HOURLY: 0.03566 × 1 node = $0.03566/hour MONTHLY: 0.03566 × 24 hours × 30.5 days × 1 node = $26.10/month*”
Read 2026-09-13 · official pricing · Report a correction
Pricing model
- Turbopuffer
- Usage-based, against a monthly minimum per plan: $16 Launch, $256 Scale, $4,096 or more for Enterprise, which the page also marks with a 35% usage premium.
- Qdrant
- Usage-based on the Standard tier: billed for compute (vCPU), memory (GB) and storage (GB) consumed by clusters, backup storage, and inference tokens for paid models.
Sources (2) →Sources ↓
- turbopuffer pricing ↗
“$16/month minimum usage [...] $256/month minimum usage [...] >= $4,096/month 35% usage premium”
Read 2026-09-14 · official pricing · Report a correction
- Qdrant Cloud Pricing ↗
“Billing is calculated based on actual resource usage during the billing period. You're charged for compute (vCPU), memory (GB), storage (GB) consumed by your clusters, storage (GB) consumed by backups, and used inference tokens of paid models.”
Read 2026-09-11 · official pricing · Report a correction
Regions
- Turbopuffer
- 20 public regions listed in the docs: 10 on AWS, 9 on GCP and 1 on Azure (eastus2); dedicated and BYOC clusters elsewhere on request.
- Qdrant
- Standard and Premium tiers run on AWS, Azure and GCP.
Sources (2) →Sources ↓
- Regions ↗
“All Regions Region Location Browser RTT aws - ap-southeast-2 Sydney, Australia [...] gcp - asia-northeast3 Seoul, South Korea [...] azure - eastus2 Virginia, US”
Read 2026-09-13 · official docs · Report a correction
- Qdrant Pricing ↗
“Cloud Providers Depends on chosen infrastructure AWS, Azure, GCP AWS, Azure, GCP AWS, Azure, GCP Uptime SLA 99.5 Single AZ: 99.9, Multi AZ: 99.95”
Read 2026-09-14 · official pricing · Report a correction
Uptime SLA
- Turbopuffer
- 99.95%, listed only against Enterprise (from $4,096/month). Launch and Scale publish no uptime SLA.
- Qdrant
- 99.5% on the paid cloud tier; 99.9% on Premium.
Sources (2) →Sources ↓
- turbopuffer pricing ↗
“Support Hours 24/7 Support SLA Uptime SLA 99.95% Contact us >= $4,096/month”
Read 2026-09-14 · official pricing · Report a correction
- Qdrant Pricing ↗
“free tokens for paid inference models 99.5% uptime SLA [...] SSO private VPC links 99.9% uptime SLA extra support talk to our engineers”
Read 2026-09-14 · official pricing · Report a correction
Max vector dimensions
- Turbopuffer
- 10,752 maximum dimensions for dense vectors; 1,024 maximum dimensions per sparse vector (sparse total dimensions unlimited)
- Qdrant
- Up to 65,535 dimensions for dense vectors
Sources (2) →Sources ↓
- Limits ↗
“Max dimensions for dense vectors 10,752 Max total dimensions for sparse vectors Unlimited (seen: 30,522) Max dimensions per sparse vector 1,024”
Read 2026-09-13 · official docs · Report a correction
- Qdrant Fundamentals - Qdrant documentation FAQ ↗
“In dense vectors, Qdrant supports up to 65,535 dimensions.”
Read 2026-09-13 · official docs · Report a correction
Similarity metrics supported
- Turbopuffer
- Cosine distance is documented as a supported metric (used in the SDK's example code); the full list of supported metrics wasn't confirmed beyond this.
- Qdrant
- Dot product, Cosine similarity (implemented as dot-product over automatically normalized vectors), Euclidean distance and Manhattan distance.
Sources (2) →Sources ↓
- Turbopuffer documentation ↗
“distance_metric="cosine_distance"”
Read 2026-09-11 · official docs · Report a correction
- Collections — Qdrant Docs ↗
“Qdrant supports these most popular types of metrics: Dot product: Dot - [wiki] Cosine similarity: Cosine - [wiki] Euclidean distance: Euclid - [wiki] Manhattan distance: Manhattan - [wiki] For search efficiency, Cosine similarity is implemented as dot-product over normalized vectors.”
Read 2026-09-14 · official docs · Report a correction
Hybrid search (vector + keyword)
- Turbopuffer
- Supported
- Qdrant
- Native hybrid queries combine dense semantic and sparse lexical (BM25) vectors in a single query
Sources (2) →Sources ↓
- Hybrid Search — turbopuffer docs ↗
“To improve search quality, multiple strategies can be used together. This is commonly referred to as hybrid search. turbopuffer supports vector search and BM25 full-text search.”
Read 2026-09-14 · official docs · Report a correction
- Combining Semantic and Lexical Search using Hybrid Search - Qdrant documentation ↗
“Hybrid search enables you to combine semantic and lexical search in a single query, returning results that match the semantic meaning, the exact keywords, or both.”
Read 2026-09-13 · official docs · Report a correction
Metadata filtering
- Turbopuffer
- Exact attribute filters on every query (and/or/not, eq, in, contains, range); the planner combines the attribute index with the ANN index.
- Qdrant
- Supported: conditions can be set on the payload and on the point id, combined with AND, OR and NOT clauses that nest arbitrarily.
Sources (2) →Sources ↓
- Query documents ↗
“Exact filters for attributes to refine search results for. Think of it as a SQL WHERE clause. [...] When combined with a vector, the query planner will automatically combine the attribute index and the approximate nearest neighbor index for best performance and recall.”
Read 2026-09-13 · official docs · Report a correction
- Filtering — Qdrant Docs ↗
“With Qdrant, you can set conditions when searching or retrieving points. For example, you can impose conditions on both the payload and the id of the point. [...] Qdrant allows you to combine conditions in clauses. Clauses are different logical operations, such as OR, AND, and NOT. Clauses can be recursively nested into each other so that you can reproduce an arbitrary boolean expression.”
Read 2026-09-14 · official docs · Report a correction
SOC 2
- Turbopuffer
- SOC 2 Type 2 audited for security and availability controls; report and penetration test available on request from the Trust Center.
- Qdrant
- SOC 2 Type 2 certified; compliance reports available on request through the Qdrant Trust Center.
Sources (2) →Sources ↓
- Security & Compliance ↗
“turbopuffer undergoes System and Organization Controls (SOC) 2 Type 2 audits of the design and operational effectiveness of security and availability controls. You can request a copy of the latest SOC 2 report and Penetration Test from our Trust Center”
Read 2026-09-13 · official docs · Report a correction
- Qdrant Security: Compliance, Isolation, and Encryption Across Every Deployment ↗
“Qdrant holds SOC 2 Type 2 and HIPAA certifications. Pull compliance reports directly from the Trust Center.”
Read 2026-09-13 · official site · Report a correction
HIPAA
- Turbopuffer
- Customers storing PHI may request a Business Associate Agreement; turbopuffer commits to the HIPAA requirements on business associates. Pricing page lists the HIPAA-ready BAA on the Scale and Enterprise plans, not on Launch.
- Qdrant
- HIPAA certified; a Business Associate Agreement is available for Qdrant Managed Cloud.
Sources (2) →Sources ↓
- Security & Compliance ↗
“Customers who wish to store protected health information (PHI) in turbopuffer may request a Business Associate Agreement (BAA) with turbopuffer, under which turbopuffer commits to compliance with the requirements of HIPAA on business associates that store and process PHI.”
Read 2026-09-13 · official docs · Report a correction
- Qdrant Security: Compliance, Isolation, and Encryption Across Every Deployment ↗
“Qdrant is HIPAA certified and Business Associate Agreement is available for Qdrant Managed Cloud.”
Read 2026-09-13 · official site · Report a correction
GDPR / data residency
- Turbopuffer
- Data stays in the region selected unless the customer moves it via the API; GDPR/CCPA Data Processing Addendum offered to all customers.
- Qdrant
- DPA offered for GDPR; cluster data is stored only in the cluster's deployment region.
Sources (2) →Sources ↓
- Security & Compliance ↗
“Customer data inserted into one region remains in that region, except when requested by the customer via the turbopuffer API. [...] turbopuffer provides a Data Processing Addendum (DPA) for all customers to enable compliance with the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA).”
Read 2026-09-13 · official docs · Report a correction
- Qdrant Security: Compliance, Isolation, and Encryption Across Every Deployment ↗
“Data in Qdrant clusters stored only in the cluster's deployment region. [...] For deployments subject to GDPR, Qdrant provides a Data Processing Agreement covering data protection and privacy commitments.”
Read 2026-09-13 · official site · Report a correction
Encryption at rest
- Turbopuffer
- Customer data always encrypted at rest with AES-256, and in transit with TLS 1.2+; optionally with a customer's own key.
- Qdrant
- All storage volumes in Qdrant Managed Cloud are encrypted at rest; data in transit is protected with TLS.
Sources (2) →Sources ↓
- Security & Compliance ↗
“Customer data and usage data is always encrypted in transit with TLS1.2+. Customer data is always encrypted at rest with AES-256”
Read 2026-09-13 · official docs · Report a correction
- Qdrant Cloud Security - Qdrant documentation ↗
“All storage volumes are encrypted at rest.”
Read 2026-09-13 · official docs · Report a correction
Customer-managed keys
- Turbopuffer
- CMEK on the Enterprise plan using keys from the customer's own KMS/EKM; the encryption key is defined at the namespace level.
- Qdrant
- Premium tier customers can bring their own encryption keys for storage volumes.
Sources (2) →Sources ↓
- Security & Compliance ↗
“turbopuffer offers support for customer managed encryption keys (CMEK), allowing customers on the Enterprise plan to ensure their data is encrypted using keys from their Key Management System (KMS)/Enterprise Key Manager (EKM). [...] the encryption key is defined at the namespace level.”
Read 2026-09-13 · official docs · Report a correction
- Qdrant Cloud Security - Qdrant documentation ↗
“Premium customers can also bring their own encryption keys for storage volumes.”
Read 2026-09-13 · official docs · Report a correction
Role-based access control
- Turbopuffer
- No built-in row or document-level RBAC; docs say permissions have to be implemented at the user level with query filters. Statement is about row/document-level RBAC; no org-level role model is documented either way.
- Qdrant
- Built-in RBAC in the Qdrant Cloud Management Console, with permissions manageable at collection level.
Sources (2) →Sources ↓
- Permissions ↗
“Permissions in turbopuffer currently have to be implemented at the user-level with filters, as turbopuffer [...] provide built-in mechanisms for row/document-level RBAC.”
Read 2026-09-13 · official docs · Report a correction
- Qdrant Security: Compliance, Isolation, and Encryption Across Every Deployment ↗
“Qdrant Cloud lets you manage permissions for cloud resources at the collection-level inside the Qdrant Cloud Management Cloud Console.”
Read 2026-09-13 · official site · Report a correction
Private networking / VPC
- Turbopuffer
- Private connections from customer VPCs to multi-tenant regions on the Enterprise plan: AWS PrivateLink, GCP Private Service Connect.
- Qdrant
- Private link connectivity from a Managed Cloud cluster to your VPC, available to Premium tier customers only.
Sources (2) →Sources ↓
- Security & Compliance ↗
“turbopuffer supports private network connections between customer VPCs and our multi-tenant regions [...] This feature is available to customers on the Enterprise plan. AWS regions use AWS PrivateLink GCP regions use GCP Private Service Connect”
Read 2026-09-13 · official docs · Report a correction
- Qdrant Security: Compliance, Isolation, and Encryption Across Every Deployment ↗
“Connect your Qdrant Managed Cloud cluster to your VPC using private links, so traffic between your application and the cluster travels over your private network. Private link connectivity is available to Premium customers.”
Read 2026-09-13 · official site · Report a correction
Dedicated infrastructure
- Turbopuffer
- Dedicated single-tenancy clusters in further regions, or BYOC clusters inside the customer's own VPC on AWS, GCP or Azure, spun up on request. Pricing page lists Single-Tenancy and BYOC under Deployment on the Enterprise plan only.
- Qdrant
- Paid Managed Cloud clusters run on their own dedicated resources, isolated in hardened, unprivileged containers.
Sources (2) →Sources ↓
- Regions ↗
“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). We can spin up dedicated or BYOC clusters in hours upon request”
Read 2026-09-13 · official docs · Report a correction
- Qdrant Cloud Security - Qdrant documentation ↗
“Paid clusters are running on their own dedicated resources to ensure stable performance and further security.”
Read 2026-09-13 · official docs · Report a correction
When these numbers change, hear about it. Sources are re-checked monthly; a repricing goes out as a short note.
Documented by only one.
These criteria are published by one provider and not the other. An absence here means we have not found a source, not that the feature is missing — except where we say we read the other provider's own page on a given date and it does not state it. Those rows lean, and the provider can correct them with a link.
Questions this comparison answers.
Entry paid plan: Turbopuffer or Qdrant?
Turbopuffer: Launch plan: $16/month minimum usage, including all database features. Scale plan: $256/month minimum. Enterprise: $4,096/month or more, plus a 35% usage premium.
Qdrant: Smallest paid Managed Cloud cluster (1 node, 0.5 vCPU, 2 GiB RAM, 8 GiB disk) on AWS us-east-1: $0.03566/hour, about $26.10/month, per Qdrant's own Cloud pricing calculator. Calculator estimate for AWS us-east-1; VAT excluded and expressly not a binding offer.
Free tier: Turbopuffer or Qdrant?
Turbopuffer: No — Entry plan (launch) carries a $16/month minimum usage; pricing lists no free plan.
Qdrant: Yes
Should I choose Turbopuffer or Qdrant?
Pick Turbopuffer if Teams that want vector search billed like storage rather than compute — object-storage-backed pricing tends to undercut memory-resident vector databases at rest, especially for large, cold datasets.
Pick Qdrant if Teams that want a choice of similarity metric rather than one fixed default, or that want the option to self-host the same open-source engine the managed cloud runs.
Comments are hosted by EchoThread, which sets cookies and offers social sign-in. Like the analytics on this site, it does not load until you accept — see the privacy page.