Competitive Pricing Built for Production
Open-source first. One pricing metric. Predictable cloud pricing for production AI memory workloads.
Context Engineering for Enterprise AI
Structured Memory Management
Store, search, and retrieve contextual information with semantic understanding. Reduce LLM hallucinations through reliable memory retrieval.
Graph-based Context Engineering
Model complex relationships between entities, memories, and context. Support n-hop traversal for deep contextual reasoning.
Cost-Optimized Intelligence
Reduce LLM token usage by 75% through efficient context retrieval. Only send relevant information to language models.
Open-core, Enterprise-ready
Self-host for control, use managed platform for scale. Same APIs everywhere—build locally, deploy globally.
Open Source
Full production stack on your infrastructure
- All core services (gateway, write, retrieve, graph)
- Unlimited ops on your own infrastructure
- Hybrid retrieval (vector + BM25 + graph)
- Knowledge graph with N-hop traversal
- Memory decay, conflict detection
- Working + procedural memory
- Python, TypeScript, LangChain, CrewAI SDKs
- MCP server included
- Docker Compose one-command deploy
- Apache 2.0 license
Builder
Fast start for production pilots and POCs
- 100K ops/month included
- 5 projects
- Unlimited end users
- Managed hosting & auto-scaling
- Transparent overage: $0.50 per 1K ops
- Email support (48h SLA)
Growth
Scaling products with operational control and SLOs
- 500K ops/month included
- 20 projects
- Admin dashboard + audit logs
- Memory consolidation + conflict resolution
- PII guardrails & data governance
- Webhook event system
- Transparent overage: $0.40 per 1K ops
- Priority support (24h SLA)
- 99.9% uptime SLA
Enterprise
Compliance, control, and high-throughput
- Committed ops volume + discounted overage
- Dedicated VPC / BYOC deployment
- SSO / SAML + SCIM provisioning
- SOC 2 / HIPAA compliance path
- Custom SLA & dedicated support
- Security reviews + architecture consulting
- Multi-region deployment
- Custom integrations & connectors
- On-prem deployment option
Simple Billing: One Unit
Knol bills cloud usage using one metric: operations (ops).
memory.add) = 1 opmemory.search) = 1 opmemory.context) = 1 opFeature Comparison
| Feature | OSS | Builder | Growth | Enterprise |
|---|---|---|---|---|
| Vector Search | ||||
| BM25 Full-text Search | ||||
| Knowledge Graph | ||||
| RRF Fusion | ||||
| PII Detection | ||||
| Multi-tenant RLS | ||||
| Python SDK | ||||
| TypeScript SDK | ||||
| Memory Decay | ||||
| Conflict Detection | ||||
| Webhook Events | ||||
| Write-time Embeddings | ||||
| N-hop Graph Traversal | ||||
| Managed Infrastructure | — | |||
| Auto-scaling | — | |||
| Memory Consolidation | ||||
| Admin Dashboard | ||||
| Audit Logging | ||||
| Custom Connectors | — | — | — | |
| SSO / SAML | — | — | — | |
| SCIM Provisioning | — | — | — | |
| Dedicated Support | — | — | — | |
| Compliance & governance packs | — | — | — | |
| Predictable overage pricing | — |
Migrate from Mem0 or Zep
Knol includes migration tooling and API-compatible patterns to reduce switching risk.
ROI Snapshot
Teams typically reduce memory-related LLM spend by optimizing context construction and avoiding redundant retrieval calls. Track three KPIs: retrieval hit rate, tokens per request, and p95 retrieval latency.
FAQ
What is context engineering?
Context engineering is the practice of structuring and managing relevant information to optimize language model outputs. Knol enables this through semantic search, knowledge graphs, and intelligent memory management—reducing hallucinations and token usage while improving reasoning accuracy.
Can I self-host everything?
Yes. Knol keeps core memory APIs, core services, SDKs, and self-host deployment open source.
How do I migrate from Mem0 or Zep?
Knol provides migration tooling, mapping checks, replay utilities, and validation reports for both Mem0 and Zep workflows. Growth and Enterprise plans include migration assistance.
What happens if I exceed my plan limits?
We notify you before usage limits are reached. Usage above included volume is billed as transparent overage per 1K ops.
Is there a free trial for paid plans?
Yes, all paid plans include a 14-day free trial with full access to all features.
What LLM is used for extraction?
The default LLM provider is configurable from the admin UI. Knol supports Gemini, Anthropic, and OpenAI-compatible providers for extraction and reasoning workflows.
Why PostgreSQL-only architecture?
PostgreSQL with pgvector and native JSON support provides exceptional reliability, security, and compliance. This eliminates vendor lock-in, simplifies self-hosting, and makes governance easier.