DeltaMemory pricing guide details
Pricing Overview
DeltaMemory uses a freemium model with four tiers: Free, Starter ($15/mo), Pro ($199/mo), and Enterprise (custom quote). The pricing scales around three core operations -- recalls, ingests, and stores -- with transparent per-operation overage rates once you exceed your plan's allocation. A yearly billing option saves roughly 12%.
We think this is a well-structured approach for an AI memory layer. The Free tier gives you enough room to prototype a single agent, and the jump to Starter at $15/mo is low enough that early-stage teams can graduate without budget pain. The real decision point is Pro at $199/mo, where you unlock unlimited memories and the Reflect/Consolidate operations that make DeltaMemory's knowledge graph genuinely useful at scale.
Plan Comparison
| Feature | Free | Starter ($15/mo) | Pro ($199/mo) | Enterprise |
|---|---|---|---|---|
| Projects | 1 | 5 | 20 | Unlimited instances |
| Memories | 3K | 50K | Unlimited | Unlimited |
| Recalls / month | 6K | 80K (then $0.002/op) | 1M (then $0.001/op) | Volume discounts |
| Ingests / month | 1K | 10K (then $0.005/op) | 100K (then $0.004/op) | Volume discounts |
| Stores / month | 2K | 30K (then $0.0005/op) | 300K (then $0.0004/op) | Volume discounts |
| API Keys | 1 | 5 | Unlimited | Unlimited |
| Backup Retention | None | 7 days | 30 days | 90+ days (custom) |
| Hibernate Timeout | 2 days | 7 days | 14 days | N/A |
| Reflect / Consolidate | No | No | 1K/mo (then $0.01/op) | Custom |
| Support | Email (72 hrs) | Email (48 hrs) | Email (24 hrs) | Dedicated CSM |
The overage model is the key detail here. Starter ingests cost $0.005 per operation once you pass 10K/mo, while Pro drops that to $0.004 with a 100K base. If your agent processes more than roughly 13K ingests per month, the Starter overages alone push you past the Pro price -- so plan accordingly.
Hidden Costs and Considerations
Watch for three cost multipliers. First, auto-hibernate: Free tier instances hibernate after just 2 days of inactivity, which means cold-start latency for intermittent workloads. Starter extends this to 7 days, Pro to 14 days. Second, Reflect and Consolidate operations are only available on Pro and above, billed at $0.01 per operation after the 1K/mo allowance -- these are essential for maintaining a healthy knowledge graph. Third, overage stacking: if you burst on recalls, ingests, and stores simultaneously, the per-operation charges compound quickly. Enterprise pricing is negotiated per deployment and includes VPC peering, dedicated nodes, and SSO/SAML.
Cost Estimates by Team Size
| Team Size | Likely Plan | Estimated Monthly Cost | Notes |
|---|---|---|---|
| Solo developer / prototype | Free | $0 | 1 project, 3K memories, sufficient for a single-agent POC |
| Small team (2-5 agents) | Starter | $15 - $40 | Base $15/mo; overages add $15-25 for moderate usage |
| Growth team (10-20 agents) | Pro | $199 - $350 | Base $199/mo; Reflect/Consolidate and overage ops add $50-150 |
| Enterprise (50+ agents) | Enterprise | Custom quote | Dedicated infrastructure, volume discounts, SLA guarantees |
How DeltaMemory Pricing Compares
DeltaMemory occupies a specific niche: persistent memory for AI agents, not general-purpose orchestration or automation. Its competitors serve adjacent but different use cases, which is reflected in their pricing structures.
| Tool | Pricing Model | Starting Price | Best For |
|---|---|---|---|
| DeltaMemory | Freemium + usage-based overages | $0 (Free) / $15/mo (Starter) | Persistent agent memory with knowledge graphs |
| Clam | Usage-based | $50/mo | Teams needing usage-based AI tooling at higher base commitment |
| LangChain | Freemium, per-seat | $0/seat (Developer) / $39/seat | Agent orchestration with LangSmith observability |
| Praes | Freemium | $24/mo (Starter) / $59/mo (Pro) | Workflow automation with tiered feature access |
DeltaMemory's $15/mo Starter plan undercuts both Praes ($24/mo) and Clam ($50/mo) at the entry level. LangChain's free developer tier is comparable for prototyping, but its $39/seat pricing adds up fast for teams -- whereas DeltaMemory charges by operations, not seats. For teams building production AI agents that need persistent memory, DeltaMemory offers the lowest entry point with the most predictable scaling curve.