Cognee

"Extract–Cognify–Load" pipeline that turns raw input into a typed, queryable knowledge graph for agent recall.

At a glance

Optimised for: typed knowledge graph extraction (ECL pipeline) over RAG

Anti-fit: no anti-fit explicitly stated

Pipeline-as-code ECL (extract/cognify/load) makes the memory build path inspectable and replayable; fully OSS.

Smaller community than Mem0/Letta; more end-to-end engineering required to deploy.

14.7k★. €7.5M seed Feb 2026. Native integrations with Claude Agent SDK and Google ADK.

API surface: REST, SDK: Python
Backend storage: hybrid (vector + graph)
Deployment: Both
Embedding model: multiple supported
Multi-tenancy: Logical namespace per project/dataset; self-hosted OSS deployment
MCP: native (first-party) — cognee-mcp
A2A: not documented publicly
OpenTelemetry: not documented publicly

Other dedicated memory layers in the catalog, ranked by inbound references.

Mem0 T1
Universal memory layer for AI agents. Three concurrent stores (vector + graph + KV); LLM-extracted facts; concurrent retrieval via ThreadPoolExecutor.
Zep & Graphiti T1
Bi-temporal knowledge graph (event time + ingestion time). Strong on chronological reasoning and contradiction tracking. Graphiti is the open-source core.
Hindsight (Vectorize) T1
Standalone memory service from Vectorize. Open source. Biomimetic four-network design (World, Bank, Observation, Opinion). Ships an MCP memory server.
Memvid T2
Single-file memory layer (one .mv2 file). No DB, no server. Append-only sequence of immutable Smart Frames with timestamps + checksums. Native Rust core (rewritten from Python).
Supermemory T1
Memory engine with API, app, browser extension, and MCP server. Extracts facts, tracks updates, resolves contradictions, auto-forgets expired info. Plugins for Claude Code, OpenCode, OpenClaw, Hermes.
AI Singapore SEA-LION T2
SEA-LION-Embedding (March 2026): retrieval + reranking models contrastively trained on 245M text pairs across 10 SE Asian languages. SEA-BED benchmark (169 datasets). SEA-LION v4 (Gemma-based) at 128K context with native function calling.

LanceDB depends on at runtime — adjacent-infrastructure cell: BYO LLM; bundles Kuzu/Neo4j for graph + LanceDB for vectors
Neo4j depends on at runtime — adjacent-infrastructure cell: BYO LLM; bundles Kuzu/Neo4j for graph + LanceDB for vectors