Postgres platform with pgvector, auth and storage.
Vespa
Search and recommendation engine with vector and tensor support.
SharkScore
#16 in Vector databases
Why it scores 63.7
Vespa has a SharkScore of 63.7/100, ranked #16 of 52 in Vector databases (updated 2026-11-19).
Each part compares Vespa with other listings in Vector databases. Evidence covers 87% of the score's weight.
- Adoption
- 36 · 30%
- User reviews
- no data · 25%
- Maintenance
- 100 · 20%
- Trust & transparency
- 80 · 15%
- Momentum
- no data · 10%
Paying never changes this score. How ranking works · method v2026-11-18.2
RankShark review of Vespa
Automated evaluation by RankShark · 18 Nov 2026 · method v1.0 · How we review
We were not paid for this review.
As of 2026-11-18, Vespa is a search and vector database platform with resource-based pricing and a $300 Vespa Cloud trial credit (E1, E2).
Vespa is a platform for building search and AI applications over vector, text, tensor, and structured data. It combines retrieval, ranking, inference, and real-time updates, and can be run as open source or through managed and supported deployment options (E1, E2).
Best for
- Teams building search or RAG applications that need to combine vector and text retrieval with ranking (E1, E2). · home page, pricing page
- Teams building recommendation, personalization, or structured-data navigation applications (E1). · home page
- Teams choosing between self-managed deployments and Vespa-managed cloud options (E2). · pricing page
Not for
- Teams that need a published fixed monthly price: the listed cloud and Enclave rates are resource-based, and some pricing is custom (E2). · pricing page
- Teams expecting an SLA with the Open Source offering: the pricing page says it has community support only and no SLA (E2). · pricing page
Key features
- Search across data types. The vendor says Vespa supports querying vectors, tensors, text, and structured data (E1). · home page
- Vector, lexical, and hybrid search. The vendor lists vector, lexical, and hybrid search, including filtered vector search (E2). · pricing page
- Ranking and inference. The vendor says the platform includes distributed machine-learned ranking and inference; its docs list ranking with TensorFlow, ONNX, XGBoost, and LightGBM models (E1, E3). · home page, docs
- Real-time updates. The vendor lists real-time indexing and partial updates (E2). · pricing page
- Deployment choices. The vendor offers self-managed Open Source, Vespa Cloud, Enclave managed in the customer's cloud, and supported self-hosted Enterprise (E2). · pricing page
Pros
- The listed capabilities include vector, lexical, and hybrid search, plus ranking and inference (E1, E2). · home page, pricing page
- The pricing page describes managed cloud, managed-in-customer-cloud, and self-managed deployment options (E2). · pricing page
- The Vespa Cloud trial provides $300 in credit and does not require a credit card (E2). · pricing page
Cons
- Cloud and Enclave costs depend on allocated resources, and the page says unit prices vary by plan; this makes the listed rates depend on workload size (E2). · pricing page
- The Open Source offering has community support only and no SLA; critical fixes arrive with the next snapshot rather than on an SLA (E2). · pricing page
- Whether customer data is used for model training is not stated on the pages we checked.
Pricing
| Plan | Price | Limits |
|---|---|---|
| Startup | Unit prices: vCPU $0.05/hour; Memory GB $0.005/hour; Disk GB $0.0002/hour; GPU GB $0.03/hour. | Community support only, with no SLA. Designed for early-stage companies developing and testing applications on Vespa Cloud. |
| Open Source | Not stated. | Full Vespa core engine, Apache 2.0, released as monthly snapshot builds. Community support only, no SLA; critical fixes arrive with the next snapshot rather than on an SLA. |
| Basic — Cloud | vCPU $0.10/hour; Memory GB $0.01/hour; Disk GB $0.0004/hour. | Initial unit pricing; non-critical production workloads. Next business day response. |
| Professional — Cloud | vCPU $0.145/hour; Memory GB $0.0145/hour; Disk GB $0.0005/hour. | Initial unit pricing; customer-facing production workloads. 24/7 support; production issues: 1-hour response; backups and DR. |
| Basic — Enclave | vCPU $0.08/hour; Memory GB $0.008/hour; Disk GB $0.0003/hour; GPU GB $0.06/hour. | Enclave is managed in your cloud. The pricing page lists these as hourly resource rates. |
| Professional — Enclave | vCPU $0.12/hour; Memory GB $0.012/hour; Disk GB $0.0004/hour; GPU GB $0.08/hour. | Enclave is managed in your cloud. The pricing page lists these as hourly resource rates. |
| Premium | Custom pricing; $20k/month minimum. | Critical workloads requiring the fastest response and named experts. 24/7 support; production issues: 15-minute response; backups and DR. |
| Enterprise | Custom pricing; annual license sized to your deployment. | Operated by your team, with Vespa support; runs in your cloud or data center. |
Checked 18 Nov 2026 · source
- Free plan
- Not stated (checked 18 Nov 2026)
- Free trial
- $300 in Vespa Cloud trial credit; sign-up without a credit card. The cluster tears down automatically when the credit runs out. · pricing page
- Trains on customer data
- Not stated (checked 18 Nov 2026)
- Data retention
- Not stated (checked 18 Nov 2026)
- Data regions
- The vendor says Enclave keeps data in the customer's cloud environment, naming AWS, Azure, and GCP. The privacy policy says personal data may be transferred outside the EU/EEA; specific Vespa Cloud regions are not listed in the captured text. · pricing page, privacy page
- Data processing agreement
- The product terms page links to a Data processing Agreement; its contents are not included in the captured page text. · terms
- AI models used
- The documentation lists using TensorFlow, ONNX, XGBoost, and LightGBM models, as well as local and external LLMs. · docs
- API
- The documentation lists query, document, state, application, config, cluster, and metrics APIs. · docs
- SDKs
- The documentation lists a Python client (pyVespa), a Java feed client, and the Vespa CLI. · docs
- Webhooks
- Not stated (checked 18 Nov 2026)
- Platforms
- The vendor lists Open Source self-managed deployment, Vespa Cloud, Enclave managed in the customer's cloud, and Enterprise supported self-hosting. Its page names AWS, Azure, and GCP for Enclave and cloud or on-premises options for some deployments. · pricing page
- Security claims
- The pricing page lists SOC 2 Type II in its security and compliance comparison; the captured text does not show which deployment options the item applies to (E2). · pricing pageThe vendor says Vespa Cloud includes built-in production monitoring and security (E2). · pricing page
What we measured
- GitHub stars · now
- 7,183
- Latest release · now
- v8.788.13 on 2026-11-16
- Package downloads · 30 days
- 392,261
- Tranco traffic rank · list of the day
- #244,861
How we tested · sources
- Hourly uptime and response-time checks of the product's URL.
- Repository, release and download data from GitHub, npm, PyPI and Hugging Face where they exist.
- We did not sign up, pay, or use the product as a customer.
- home page: https://vespa.ai/ · fetched 18 Nov 2026
- pricing page: https://vespa.ai/pricing/ · fetched 18 Nov 2026
- docs: https://docs.vespa.ai/ · fetched 18 Nov 2026
- privacy page: https://vespa.ai/privacy-policy/ · fetched 18 Nov 2026
- terms: https://vespa.ai/product-terms/ · fetched 18 Nov 2026
- security page: https://trust.vespa.ai/ · fetched 18 Nov 2026
Something wrong? Report an error with a public link that shows it; corrections re-run the review.
Top alternatives to Vespa
See all- 78.9ToolFreemium · from $25/mo
- 75.3
In-memory data store with vector search for AI apps.
ToolFreemium · from $5/mo - 73.7
Open-source search and analytics with a vector engine.
ToolOpen source - 73.7
Vector similarity search for Postgres.
ToolOpen source