MongoDB Atlas Vector Search Pricing 2026
Complete pricing guide with plans, hidden costs, and cost analysis
MongoDB Atlas Vector Search has a free plan. Paid plans start at $8/per hour (Flex) and go up to $33.26/per hour.
MongoDB Atlas Vector Search costs Free to $33.26 per per hour as of April 2026, with 4 plans available including a free tier. Plans: M0 Free (free), Shared (M2/M5) at $9/per hour, and Flex at $8/per hour. Pricing depends on your chosen tier, contract length, and negotiated discounts.
Use the interactive pricing calculator to estimate your exact cost based on team size and requirements.
- Free tier: Yes
MongoDB Atlas Vector Search offers 4 pricing tiers: M0 Free, Shared (M2/M5), Flex, Dedicated (M10+). A free plan is available. Paid plans include Shared (M2/M5) at $9/month, Flex at $8/month. The Shared (M2/M5) plan is very small production apps.
Compared to other vector databases software, MongoDB Atlas Vector Search is positioned at the budget-friendly price point.
- 2 documented hidden costs beyond list price
How much does MongoDB Atlas Vector Search cost?
MongoDB Atlas Vector Search Pricing Overview
MongoDB Atlas Vector Search has 4 pricing plans, including a free tier. Paid plans range from $0 to $33.26/per hour. The M0 Free plan is free and is best for learning, prototypes, and exploration. The Shared (M2/M5) plan costs $9/per hour, best for very small production apps. The Flex plan costs $8/per hour, best for apps with unpredictable or low traffic. The Dedicated (M10+) plan requires contacting sales for a custom quote and is designed for production workloads of any scale.
There are at least 2 documented hidden costs beyond MongoDB Atlas Vector Search's list price, including implementation, training, and add-on fees.
This pricing was last verified in April 23, 2026.
MongoDB Atlas Vector Search pricing starts at $0/month on the M0 Free tier, which provides a sandboxed shared cluster for development and prototyping. For low-traffic workloads, the Flex tier starts at $8/month and the Shared (M2/M5) tier starts at $9/month, both on shared infrastructure with no dedicated compute. Production applications requiring dedicated resources and uptime SLAs move to Dedicated (M10+) clusters, which are custom-priced based on cluster configuration and region.
How MongoDB Atlas Vector Search Pricing Compares
Compare MongoDB Atlas Vector Search pricing against top alternatives in Vector Databases.
All MongoDB Atlas Vector Search Plans & Pricing
| Plan | Monthly | Annual | Best For |
|---|---|---|---|
| M0 Free | Free | Free | Learning, prototypes, and exploration |
| Shared (M2/M5) | $9 /month | Custom | Very small production apps |
| Flex | $8 /month | Custom | Apps with unpredictable or low traffic |
| Dedicated (M10+) | Custom | Custom | Production workloads of any scale |
View all features by plan
M0 Free
- Free forever
- 512 MB storage
- Shared RAM and vCPU
- Vector Search included at no additional cost
Shared (M2/M5)
- M2: $9/month (2 GB storage)
- M5: $25/month (5 GB storage)
- Shared cluster infrastructure
- Vector Search included
Flex
- $0.011/hour base
- Billed $8-$30/month based on ops/sec
- Auto-scales with operations
- Vector Search included
Dedicated (M10+)
- Hourly billing by cluster tier
- 2 GB to 768 GB RAM
- 2 to 96 vCPUs
- Vector Search included; optional S-series Search Nodes for workload isolation
Usage-Based Rates
Per-unit pricing for MongoDB Atlas Vector Search API usage.
Flex
| Model | Unit | Rate |
|---|---|---|
| flex-cluster | hour | $0.011 |
- Billing bands: 0-100 ops/sec $8/mo; 100-200 $15/mo; 200-300 $21/mo; 300-400 $26/mo; 400-500 $30/mo
Dedicated (M10+)
| Model | Unit | Rate |
|---|---|---|
| m10 | hour | $0.080 |
| m20 | hour | $0.200 |
| m30 | hour | $0.540 |
| m40 | hour | $1.04 |
| m50 | hour | $2.00 |
| m60 | hour | $3.95 |
| m80 | hour | $7.30 |
| m140 | hour | $10.99 |
| m200 | hour | $14.59 |
| m300 | hour | $21.85 |
| m700 | hour | $33.26 |
| s20 | hour | $0.120 |
| s30 | hour | $0.240 |
| s40 | hour | $0.480 |
| s50 | hour | $0.990 |
| s60 | hour | $1.77 |
| s70 | hour | $2.50 |
| s80 | hour | $3.26 |
- M-series are primary cluster nodes; S-series are high-CPU Search Nodes for dedicated vector/search workloads
- Storage and egress billed separately by cloud provider
Compare MongoDB Atlas Vector Search vs Alternatives
Before committing to MongoDB Atlas Vector Search, compare pricing with these 3 alternatives in the same category.
Prototyping and small projects
Full comparisonTesting and evaluation
Full comparisonTesting and prototypes
Full comparisonAll MongoDB Atlas Vector Search alternatives & migration guides
What Companies Actually Pay for MongoDB Atlas Vector Search
MongoDB Atlas Vector Search Year 1 Total Cost by Company Size
Real deployment costs including licenses, implementation, training, and admin — not just the sticker price.
Single developer or student using the M0 Free shared cluster for a prototype or small project within free tier limits.
Small team using the Flex plan for variable or low-traffic workloads.
Small team using a Shared (M2/M5) cluster for a low-traffic application.
A solo developer running a production-adjacent workload on a dedicated Atlas cluster; one Reddit commenter reports spending around $100/month in this configuration.
A startup ingesting 8–10 GB of structured JSON data per day requiring a dedicated M10 cluster for consistent performance
CURRENT TIER DATA
How MongoDB Atlas Vector Search Pricing Compares
| Software | Starting Price | Top Price |
|---|---|---|
| MongoDB Atlas Vector Search | Free | $33.26/per hour |
| Chroma | Free | $250/month |
| LanceDB | Free | $1000/month |
| Milvus | Free | $155/month |
| pgvector | Custom | Custom |
| Pinecone | Free | $500/month |
Detailed pricing comparisons:
MongoDB Atlas Vector Search Pricing FAQ
01 Is MongoDB Atlas Vector Search free?
Yes, MongoDB Atlas Vector Search is available on the M0 Free tier at $0/month, suitable for development and hobby projects. Paid tiers start at $8/month (Flex) or $9/month (Shared M2/M5) for small teams, with Dedicated clusters (M10+) priced based on compute and storage configuration.
02 Are shared Atlas clusters secure enough for real projects?
Yes. Shared instances are sandboxed and secure for most hobby and small projects. As one MongoDB community member notes: 'shared instances are still secure. While theoretically possible to exploit the fact that the instance is shared, it isn't feasible since the user does not have direct access. Furthermore, shared instances are sandboxed.' Dedicated clusters add extra security features but are not required for most non-enterprise workloads.
03 When does MongoDB Atlas pricing become expensive?
Costs tend to rise when moving from shared or Flex plans to Dedicated (M10+) clusters for production workloads. The M0 Free and Shared (M2/M5 at $9/month) tiers cover most development needs, but dedicated configurations can reach approximately $100/month or more for a solo developer, and costs scale with compute and storage requirements.
04 Does MongoDB have a startup program?
Yes. MongoDB is reported to have startup credit programs offering $1,000–$2,000 in credits for eligible early-stage startups, similar to programs offered by AWS and Google.
05 What is the cheapest paid tier for MongoDB Atlas Vector Search?
The Flex tier starts at $8/month and the Shared (M2/M5) tier starts at $9/month. Both run on shared infrastructure without dedicated compute. For production workloads requiring dedicated resources, you would need a Dedicated (M10+) cluster, which is custom-priced.
06 How much does a dedicated MongoDB Atlas cluster cost for production vector search?
Dedicated (M10+) cluster pricing is custom and varies by configuration. Community reports suggest M10 clusters can cost $600–$700/month for data-intensive startup workloads. MongoDB's pricing calculator should be consulted for your specific cluster and region requirements.
07 What are the limitations of the free M0 tier for vector search?
The M0 Free tier uses a shared cluster with no guaranteed performance, no SLA, and fewer security features than dedicated clusters. There is no VPC peering, private endpoints, or encryption at rest. Shared compute means query performance can degrade when other tenants are active.
Is this pricing incorrect? — we'll verify and update it.