Cognee vs Supermemory
Cognee vs Supermemory: side-by-side comparison of two dedicated memory layers systems — architecture, taxonomy, license, pricing, MCP/A2A support, and direct edges.
Cost & capability
| Cognee | Supermemory | |
|---|---|---|
| Capability band | competent | competent |
| Capability composite | 65 | 72 |
| Cost tier | free | free |
| $/Mtok input | 0 | 0 |
| $/Mtok output | 0 | 0 |
| Use cases | Long Running Session, Memory Augmented Chat, Analytical Summarization, Multi Agent Coordination | Long Running Session, Memory Augmented Chat, Code Generation Focused |
Where they differ (14)
Rows where both sides have data and the values disagree — the shortlist of dimensions that actually distinguish these two systems.
| Cognee | Supermemory | |
|---|---|---|
| Capability composite | 65 | 72 |
| Use cases | Long Running Session, Memory Augmented Chat, Analytical Summarization, Multi Agent Coordination | Long Running Session, Memory Augmented Chat, Code Generation Focused |
| Type | Knowledge graph + ECL pipeline | Memory graph + extraction + RAG |
| Created | 2023-08 | 2024-02 |
| Latest release | v1.0.8 2026-05-06 | no releases |
| License | Apache-2.0 | MIT |
| GitHub | 17.1k★ +152/mo Python | 22.4k★ +113/mo TypeScript |
| Pricing | Free + paid | OSS |
| Funding | $10M total Seed · 2026-02 | $6M total Seed · 2025-10 |
| Backend storage | hybrid (vector + graph) | custom |
| API surface | REST, SDK: Python | REST, SDK: Python, JS/TS |
| Multi-tenancy | Logical namespace per project/dataset; self-hosted OSS deployment | Enterprise: deploy in customer VPC for full tenant isolation; standard SaaS uses logical namespace per user/org |
| MCP | native (first-party) — cognee-mcp | native (first-party) — claude-supermemory plugin + MCP |
| Optimised for | typed knowledge graph extraction (ECL pipeline) over RAG | multi-channel capture (API, app, browser ext, MCP) + RAG over personal graph |
At a glance
| Cognee | Supermemory | |
|---|---|---|
| Section | Dedicated memory layers | Dedicated memory layers |
| Tier | T1 | T1 |
| Type | Knowledge graph + ECL pipeline | Memory graph + extraction + RAG |
| Created | 2023-08 | 2024-02 |
| Latest release | v1.0.8 2026-05-06 | no releases |
| License | Apache-2.0 | MIT |
| GitHub | 17.1k★ +152/mo Python | 22.4k★ +113/mo TypeScript |
| Pricing | Free + paid | OSS |
| Funding | $10M total Seed · 2026-02 | $6M total Seed · 2025-10 |
| Backend storage | hybrid (vector + graph) | custom |
| Deployment | Both | Both |
| API surface | REST, SDK: Python | REST, SDK: Python, JS/TS |
| Embedding | multiple supported | multiple supported |
| Multi-tenancy | Logical namespace per project/dataset; self-hosted OSS deployment | Enterprise: deploy in customer VPC for full tenant isolation; standard SaaS uses logical namespace per user/org |
| MCP | native (first-party) — cognee-mcp | native (first-party) — claude-supermemory plugin + MCP |
| A2A | not documented publicly | not documented publicly |
| OpenTelemetry | not documented publicly | not documented publicly |
| Optimised for | typed knowledge graph extraction (ECL pipeline) over RAG | multi-channel capture (API, app, browser ext, MCP) + RAG over personal graph |
| Anti-fit | no anti-fit explicitly stated | no anti-fit explicitly stated |
Taxonomy
| Axis | Cognee | Supermemory |
|---|---|---|
| storage | graph | graph |
| retrieval | graph-traversal | similarity |
| persistence | long-term | long-term |
| update | extraction | extraction |
| unit | fact | document |
| governance | inspectable | opaque |
| conflict | llm-arbitrate | llm-arbitrate |
Pros & cons
Cognee
Pros: Pipeline-as-code ECL (extract/cognify/load) makes the memory build path inspectable and replayable; fully OSS.
Cons: Smaller community than Mem0/Letta; more end-to-end engineering required to deploy.
Supermemory
Pros: Simple universal API surface — wraps memory in a single SDK call without forcing extraction/retrieval choices; YC-backed.
Cons: Architectural opacity is the cost of simplicity — limited control over structure or eviction; small team.