AI Databases & Vector Stores
ApertureData pricing: what do the Basic ($0.33/hr), Standard ($1.29/hr), and Premium ($4.00/hr) plans include, and which fits a production pilot?
7 min read
ApertureDB Cloud pricing is structured so you can move from experimentation to production pilots without re-architecting your data layer. The Basic, Standard, and Premium tiers map to distinct stages of multimodal AI maturity—what changes across them is not just CPU/RAM, but the operational guarantees you get for real workloads.
Quick Answer: Basic ($0.33/hr) is ideal for small projects and early PoCs, Standard ($1.29/hr) is the default for serious production pilots and most production apps, and Premium ($4.00/hr) is designed for high‑throughput, high‑reliability production at scale with more replicas and support. If you’re running a production pilot for multimodal RAG, GraphRAG, or agents, Standard is usually the best fit.
Frequently Asked Questions
What does the Basic plan at $0.33/hr actually include?
Short Answer: Basic includes a production-grade ApertureDB Cloud instance with 2 vCPUs, 8 GB RAM, 64 GB storage, SSL encryption, RBAC, and Basic Support—sized for small projects, PoCs, and low-traffic apps.
Expanded Explanation:
The Basic tier is the on-ramp from experimentation to something that looks like a real application. You get ApertureDB as the same “vector + graph + multimodal storage” engine we run for larger customers—just on a smaller footprint and with non-enterprise support.
It’s ideal when you’re still proving out multimodal pipelines (images, videos, documents, text, audio, embeddings, metadata) and don’t yet have heavy QPS or strict uptime expectations. Many teams use Basic to prototype their retrieval flows (RAG, GraphRAG, dataset preparation) and validate query patterns like “vector search + metadata filters + graph traversal in a single call” before scaling out.
Key Takeaways:
- 2 CPU, 8 GB RAM, 64 GB storage, Basic Support
- Best for PoCs, small workloads, and early-stage multimodal experimentation
How do I move from free trial to a paid plan without breaking my pipelines?
Short Answer: You can start with the 30-day free ApertureDB Cloud trial, then upgrade into Basic, Standard, or Premium without changing your application code or query patterns.
Expanded Explanation:
The trial environment runs the same core ApertureDB engine and AQL query interface as the paid tiers. That means your ingestion workflows (Ingest Dataset, Generate Embeddings, Detect Faces and Objects) and your retrieval logic (vector search + filters + graph traversals) carry over unchanged. Upgrading is about resources, SLAs, and support—not about replatforming.
Operationally, you go from a time-bound environment to a fully managed instance with more predictable performance, storage capacity, and support guarantees. From there, you can scale vertically (bigger instance class) and, in Standard/Premium, add replicas for higher availability and read throughput.
Steps:
- Use the free 30-day trial to stand up your multimodal memory layer and validate queries.
- Monitor resource usage (CPU, RAM, storage, QPS) as you run real workloads.
- Upgrade to Basic, Standard, or Premium based on your production pilot’s expected load and reliability needs—without changing your application logic.
How do Basic ($0.33/hr), Standard ($1.29/hr), and Premium ($4.00/hr) compare?
Short Answer: Basic is small and budget-friendly, Standard steps up to a production-ready configuration with more RAM/CPU and a replica, and Premium adds larger resources, more replicas, and stronger support for mission-critical traffic.
Expanded Explanation:
All three plans give you the same foundational capabilities: multimodal storage (images, videos, docs, text, audio), high-performance vector search, and property-graph queries in one database. Where they diverge is in capacity, redundancy, and support—things that matter directly to your on-call experience and SLOs.
- Basic: designed for smaller or early-stage applications where occasional contention or capacity constraints are acceptable.
- Standard: sized for production apps that need low latency and consistent performance as query volume grows, with one replica to offload reads and improve availability.
- Premium: built for teams expecting high QPS, larger datasets, and stronger uptime requirements; you get more resources, more replicas, and premium support to keep your vector + graph layer stable at scale.
Comparison Snapshot:
- Basic ($0.33/hr): 2 CPU, 8 GB RAM, 64 GB storage, Basic Support — best for PoCs and low-volume apps.
- Standard ($1.29/hr): 8 CPU, 32 GB RAM, 512 GB storage, 1 replica, Standard Support — best default choice for production pilots and most production workloads.
- Premium ($4.00/hr): 10 CPU, 48 GB RAM, 1 TB storage, 2 replicas, Premium Support — best for high-traffic, high-reliability production systems.
Which ApertureDB plan is best for a production pilot?
Short Answer: For most teams, the Standard plan ($1.29/hr) is the right starting point for a production pilot—it gives you enough CPU/RAM, a replica, and support to test real workloads without overprovisioning.
Expanded Explanation:
A credible production pilot needs to mimic real-world conditions: higher concurrency, varied queries, and realistic dataset sizes spanning images, videos, documents, and embeddings. Standard is designed for exactly this: 8 vCPUs, 32 GB RAM, 512 GB storage, and one replica for better read performance and resilience.
With Standard, you can push ApertureDB closer to the workloads we highlight publicly—sub-10ms vector search latency, 13K+ queries/sec, and billion-scale graph traversal (~15ms lookup)—without hitting resource ceilings too early. You also get Standard Support, which matters once you’re integrating agents, RAG pipelines, and internal tools that a real team depends on.
If your pilot already expects very high QPS or needs stronger redundancy from day one, jump straight to Premium. But if you’re coming from text-only vector stores or brittle “object store + embeddings + SQL” stacks, Standard is usually the right balance of cost and operational realism.
What You Need:
- A clear sense of expected QPS, dataset size (media + embeddings + metadata), and latency targets
- A willingness to run realistic load tests (RAG, GraphRAG, agent memory queries) on Standard before deciding whether to scale to Premium
How should I think about pricing strategically for multimodal RAG and agent workloads?
**Short Answer:**Optimize for predictable TCO and reliability, not just hourly rate—Standard or Premium can reduce overall cost by eliminating separate vector, graph, and storage systems and the pipelines between them.
Expanded Explanation:
Most multimodal AI projects burn time and money on glue code: stitching together object storage, a vector DB, a relational store, and a graph engine. Each system adds its own infra bill plus hidden costs—pipeline maintenance, schema migrations, on-call overhead when integrations fail.
ApertureDB’s pricing needs to be viewed in that context. Basic, Standard, and Premium are not just bigger machines; they’re “one database, many applications” that let you consolidate media storage, embeddings, metadata, and graph relationships into a single operational system. That’s how customers report 2.5x query speedups in production and move from unstable 4K QPS stacks to >10K QPS with high stability.
If your goal is to move from prototype to production 10× faster and save 6–9 months of infrastructure setup, it usually makes sense to run your pilot on Standard (or Premium if you’re aggressive) rather than under-provisioning and compensating with more complexity elsewhere.
Why It Matters:
- Consolidating vector + graph + multimodal storage into one system lowers integration costs and reduces on-call risk.
- Picking the right tier up front (typically Standard for pilots) gives you a realistic view of performance and reliability before you scale users or agents.
Quick Recap
Basic, Standard, and Premium all run the same ApertureDB engine—the foundational data layer that unifies multimodal storage, vector search, and graph traversal. Basic ($0.33/hr) is tuned for PoCs and low-volume workloads, Standard ($1.29/hr) is the default choice for production pilots and most production systems, and Premium ($4.00/hr) is built for high-throughput, high-availability deployments. For a production pilot that needs credible performance and reliability signals, Standard is usually the right starting point.