Memory your agents can afford. Predictable pricing.
Entity-scoped memory, semantic recall, source-backed retrieval, and storage in a single API. Stop paying separately for S3 + Pinecone + embedding APIs.
Free
Solo NodeEverything you need to give one prototype agent durable memory.
- Remembers up to 500 facts or notes
- 100K indexed memory chunks
- 10K recalls / month
- 1 GB memory storage
- Python, TypeScript, Go, .NET SDKs
- Community support
Pro
MeshEntity-scoped memory for teams shipping AI-native products.
- Remembers up to 5,000 facts or notes
- 1M indexed memory chunks
- 100K recalls / month
- 25 entity partitions
- 15 GB memory storage
- Replication & sharding
Scale
ClusterHigh-volume memory for distributed agent fleets.
- Remembers up to 25,000 facts or notes
- 5M indexed memory chunks
- 500K recalls / month
- 1,000 entity partitions
- Backup snapshots
- Isolation proof for tenants
Business
ConstellationMemory platform for teams that need security, compliance, and priority support.
- Remembers up to 100,000 facts or notes
- 15M indexed memory chunks
- 2M recalls / month
- 10,000 entity partitions
- 150 GB memory storage
- SSO, RBAC & audit log
Replace 3 services
Storage + vectors + embeddings in one API. No S3, no Pinecone, no embedding API bills.
Your data, isolated
Encrypted, single-tenant storage on Aether's infrastructure — export or delete your data anytime.
Scales with you
From your first request to high throughput, scaling and redundancy are handled for you.
The Flat-Rate Advantage
One predictable price, not four metered ones
A self-assembled stack meters you for vector reads, storage egress, embedding tokens, and per-search reranking — costs that climb with every successful query. Aether bundles all four into one flat rate that doesn't move with usage. Figures below derive from dated 2026 list pricing for each plan's workload.
Pro
Save 81%Same workload on a metered Pinecone + S3 + embeddings + rerank stack.
Scale
Save 81%Same workload on a metered Pinecone + S3 + embeddings + rerank stack.
Business
Save 88%Same workload on a metered Pinecone + S3 + embeddings + rerank stack.
Tired of unpredictable cloud bills?
See exactly how much you're overpaying for vector compute, storage egress, and token markup. Calculate the true ROI of Aether's flat-rate architecture.
Compare every feature
Predictable baseline pricing with clear usage-based overages beyond included capacity. Every plan includes all SDKs, all LLM integrations, and local ONNX embeddings.
Feature comparison
Free
Everything you need to start building RAG pipelines locally.
Capacity
- Documents
- 500
- Vectors
- 100K
- Queries / month
- 10K
- Storage
- 1 GB
- Replication factor
- 1x
- API keys
- 2
- Archive bytes / month
- Not included
Infrastructure
- Multi-node replication
- No
- Data sharding
- No
- Federated search
- No
- Prometheus metrics
- No
- Backup snapshots
- No
- Custom embedding models
- No
- Permanent archive (Arweave)
- No
Security & Compliance
- RBAC
- No
- SSO (SAML / OIDC)
- No
- Audit log
- No
- Uptime SLA
- None
Support
- Community (GitHub / Discord)
- Yes
- Email support
- No
Pro
Replicated clusters for teams shipping AI-native products.
Capacity
- Documents
- 5,000
- Vectors
- 1M
- Queries / month
- 100K
- Storage
- 15 GB
- Replication factor
- 2x
- API keys
- 10
- Archive bytes / month
- 1 GB
Infrastructure
- Multi-node replication
- Yes
- Data sharding
- Yes
- Federated search
- Yes
- Prometheus metrics
- Yes
- Backup snapshots
- No
- Custom embedding models
- No
- Permanent archive (Arweave)
- No
Security & Compliance
- RBAC
- No
- SSO (SAML / OIDC)
- No
- Audit log
- No
- Uptime SLA
- None
Support
- Community (GitHub / Discord)
- Yes
- Email support
- (48hr SLA)
Scale
Expanded capacity for globally distributed agent fleets.
Capacity
- Documents
- 25,000
- Vectors
- 5M
- Queries / month
- 500K
- Storage
- 50 GB
- Replication factor
- 2x
- API keys
- 50
- Archive bytes / month
- 50 GB
Infrastructure
- Multi-node replication
- Yes
- Data sharding
- Yes
- Federated search
- Yes
- Prometheus metrics
- Yes
- Backup snapshots
- Yes
- Custom embedding models
- No
- Permanent archive (Arweave)
- No
Security & Compliance
- RBAC
- No
- SSO (SAML / OIDC)
- No
- Audit log
- No
- Uptime SLA
- 99.9%
Support
- Community (GitHub / Discord)
- Yes
- Email support
- (24hr SLA)
Business
Full platform with security, compliance, and priority support.
Capacity
- Documents
- 100,000
- Vectors
- 15M
- Queries / month
- 2M
- Storage
- 150 GB
- Replication factor
- 3x
- API keys
- Unlimited
- Archive bytes / month
- 100 GB
Infrastructure
- Multi-node replication
- Yes
- Data sharding
- Yes
- Federated search
- Yes
- Prometheus metrics
- Yes
- Backup snapshots
- Yes
- Custom embedding models
- Yes
- Permanent archive (Arweave)
- No
Security & Compliance
- RBAC
- Yes
- SSO (SAML / OIDC)
- Yes
- Audit log
- Yes
- Uptime SLA
- 99.95%
Support
- Community (GitHub / Discord)
- Yes
- Email support
- (4hr SLA)
Feature comparison
Capacity
| Feature | Free tier | Pro tier | Scale tier | Business tier |
|---|---|---|---|---|
| Documents | 500 | 5,000 | 25,000 | 100,000 |
| Vectors | 100K | 1M | 5M | 15M |
| Queries / month | 10K | 100K | 500K | 2M |
| Storage | 1 GB | 15 GB | 50 GB | 150 GB |
| Replication factor | 1x | 2x | 2x | 3x |
| API keys | 2 | 10 | 50 | Unlimited |
| Archive bytes / month | Not included | 1 GB | 50 GB | 100 GB |
Infrastructure
| Feature | Free tier | Pro tier | Scale tier | Business tier |
|---|---|---|---|---|
| Multi-node replication | No | Yes | Yes | Yes |
| Data sharding | No | Yes | Yes | Yes |
| Federated search | No | Yes | Yes | Yes |
| Prometheus metrics | No | Yes | Yes | Yes |
| Backup snapshots | No | No | Yes | Yes |
| Custom embedding models | No | No | No | Yes |
| Permanent archive (Arweave) | No | No | No | No |
Security & Compliance
| Feature | Free tier | Pro tier | Scale tier | Business tier |
|---|---|---|---|---|
| RBAC | No | No | No | Yes |
| SSO (SAML / OIDC) | No | No | No | Yes |
| Audit log | No | No | No | Yes |
| Uptime SLA | None | None | 99.9% | 99.95% |
Support
| Feature | Free tier | Pro tier | Scale tier | Business tier |
|---|---|---|---|---|
| Community (GitHub / Discord) | Yes | Yes | Yes | Yes |
| Email support | No | (48hr SLA) | (24hr SLA) | (4hr SLA) |
Need hyperscale or enterprise controls?
Our Enterprise Sovereign tier offers unlimited scaling, strict tenant isolation, and custom embedding models.
Pricing questions
The same plan source powers pricing, billing, and dashboard limits, so the numbers stay aligned across Aether.
What happens when I hit a plan limit?
Free-tier hard limits return 402 with a plan-limit error code. Paid plans include a grace buffer for annual billing and then bill overages where the plan supports them. Rate limits return 429 with a retry-after header so clients can back off cleanly.
How do upgrades and downgrades work?
You can change plans from the billing page in the Aether dashboard. Upgrades take effect immediately. Downgrades apply at the next billing period and remain available only when current usage fits inside the target plan's limits.
What is the refund policy?
Monthly and annual subscriptions are cancellable from the dashboard. Refund requests are reviewed case by case for duplicate charges, accidental upgrades, and service-impacting incidents.
When should I talk to sales?
Talk to sales for Enterprise, dedicated environments, compliance reviews, custom embedding models, high-volume migration support, or limits beyond the published plans.
Start building with Aether today.
Free forever on the Solo Node plan. No credit card required. Upgrade when you need replication, redundancy, and enterprise features.
© 2026 Quintessence Group, Inc. All rights reserved.