Vespa Cloud

Managed Vespa, the open-source big-data search and ranking engine spun out of Yahoo. Billed per vCPU, memory and disk hour.

The smallest production setup modeled here costs about $695/month. Vespa bills for CPU, memory and disk hours, with at least two nodes per production cluster and 8 GB of memory per node. Heavy ranking or vector search may need more than this example. Volume and annual-commitment discounts are available. The company is Norwegian, with EU deployments on AWS and GCP.

Headquarters: 🇳🇴 Norway
Parent jurisdiction: 🇳🇴 Norway
EU self-serve plan: Available; check eligible plans and regions
GDPR DPA: Yes
Billing: payg
Free tier: Free trial with $300 in credits; Startup plan for dev zones only.
Open source: Yes · self-hostable
Website: vespa.ai
Source last checked: 2026-10-06
Pricing evidence: Published rates

Pricing

Calculated from the listed vendor rates; exclusions still apply. How to read these estimates.

Plan	Base fee / month
Basic, minimal production (2 container + 2 content nodes, 8 GB)	$694.96
Basic, 2 container + 2 content nodes (32 GB)	$1,525.41

Allowances, plan limits & calculation details

Rates are shown at their recorded precision. A zero means no charge in this calculation; the vendor may charge for things we leave out. See the notes for usage assumptions and currency conversions.

Basic, minimal production (2 container + 2 content nodes, 8 GB)

Base fee / month: $694.96
Included records (thousands): 0
1k additional records / month: $0
Included searches / month (thousands): 0
1k additional searches: $0
Modeled plan limits: ≤ 5,000 records (thousands)

Basic, 2 container + 2 content nodes (32 GB)

Base fee / month: $1,525.41
Included records (thousands): 0
1k additional records / month: $0
Included searches / month (thousands): 0
1k additional searches: $0
Modeled plan limits: ≤ 25,000 records (thousands)

Basic plan prices: $0.10 per vCPU-hour, $0.01 per GB memory-hour and $0.0004 per GB disk-hour (Commercial $0.145/$0.0145/$0.0005, Enterprise $0.18/$0.018/$0.0007 with a $20k/month minimum; Startup at half of Basic is for dev zones only). Production needs at least 2 nodes per cluster and 8 GB memory per node. The minimal setup modelled is 2 container nodes (1 vCPU, 8 GB, 16 GB disk) plus 2 content nodes (2 vCPU, 8 GB, 24 GB disk), 730 h/month = about $695. For ~20M records at 1 KB, the content nodes go to 4 vCPU / 32 GB / 96 GB disk, about $1,525. These are rough sizings; heavy ranking or vector search needs more. Prices drop with volume (up to 50%), and annual commitment saves 15%. EU zones: AWS Ireland, London, Frankfurt, Stockholm and GCP europe-west3. Vespa.ai AS, Trondheim.

Managed alternatives

At 1M records, 5M searches / month. Adjust usage and requirements.

Vespa Cloud: $8,340/yr estimated.

Bonsai: $600/yr estimated · $7,740/yr bill differenceModeled ratesAssumes roughly 1 KB per record and checks document counts, not concurrency. Index overhead, peak traffic and availability can require a larger plan.
Typesense Cloud: $864/yr estimated · $7,476/yr bill differenceModeled ratesSingle-node cluster with 1 KB records and modeled sizing. Production HA roughly triples cost; EU regions can cost more.
Amazon OpenSearch Service: $1,063/yr estimated · $7,276/yr bill differenceModeled ratesSingle-node domain sized at 1 KB per record. Multi-zone availability, replicas and peak query load are not modeled.
Elastic Cloud (Elasticsearch): $2,520/yr estimated · $5,820/yr bill differenceModeled ratesUses published example deployments and assumes 1 KB records. Real query load and resource sizing can differ.

The price difference leaves out migration costs, operating costs and any features you’d need to replace.

Self-hosted options

Operating cost is not estimated. These options are not counted as savings against managed services.

Other search comparisons

Found a wrong rate or plan restriction? Report a correction with a current source.