Open sourcePython · ≈33k stars · Apache-2.0
Graphiti
Temporal knowledge graphs for agent memory.
Graphiti (by Zep) builds dynamically updating, time-aware knowledge graphs from unstructured data — powering agentic memory that knows not just facts but when they were true.
Pros & cons
Each point cites its sources and says what kind of evidence backs it. A second reviewer tried to refute every claim before it was kept.
Pros
- DocsBuilds a temporal knowledge graph that takes in new episodes incrementally without batch recomputation, so agents can track facts that change over time.[1]
- DocsApache-2.0, and you can bring your own graph store: Neo4j 5.26, FalkorDB, or Amazon Neptune with OpenSearch.[1]
- MetadataActively maintained, with v0.31.2 released 2026-10-20 and about 32k stars.[1]
Cons
- DocsSelf-hosting means running a graph database: Kuzu support is deprecated after KuzuDB was archived, and the README lists no Postgres/pgvector backend even though users asked for one.[1][2][3]
- DocsIngestion runs LLM extraction and defaults to OpenAI for both LLM and embeddings; SEMAPHORE_LIMIT defaults to 10 concurrent operations to avoid provider 429s, which slows large imports.[1]
- DocsThe README warns that providers without Structured Output, especially smaller models, cause schema errors and ingestion failures, and a pydantic ExtractedEntities validation bug with a local model is still open.[1][2]
- User reportsAn infinite-loop risk in label propagation (no maximum iteration cap) has been open since April 2025.[1]
- Best for
- Teams building their own agent memory that needs time-aware facts and can run Neo4j or FalkorDB themselves.
- Watch out for
- Per-episode LLM cost and provider rate limits slow ingestion, so budget for them before indexing large histories.
Assessed Oct 29, 2026 · adversarially reviewedSuggest a correction
Activity
- Last push
- · yesterday
- Latest release
- v0.31.2
· 4w ago - Last 90 days
- 6 releases
v0.31.2v0.31.2 - update certifi to mitigate eMudhra compromise
· 4w agoWhat's Changed chore(mcp): require graphiti-core 0.31.1 and set MCP version to 1.1.1 by @devin-ai-integration[bot] in https://github.com/getzep/graphiti/pull/2013 chore(ci): add the Devin bot to the CLA allowlist by @danielchalef in https://github.com/getzep/graphiti/pull/2028…
- · 5w ago
What's Changed chore: set graphiti-core version to 0.31.0 for the v0.31.0 release by @devin-ai-integration[bot] in https://github.com/getzep/graphiti/pull/2010 chore: set graphiti-core version to 0.31.1 by @devin-ai-integration[bot] in…
v0.31.0v0.31.0 - gpt-6-luna as the default OpenAI model
· 5w agoWhat's Changed Publish zepai/graphiti on v.. tags like MCP by @mehulp93 in https://github.com/getzep/graphiti/pull/1862 Add pevans to the CLA allowlist [ZEPAI-3570] by @pevans in https://github.com/getzep/graphiti/pull/1883 Update root uv.lock to fix Dependabot alerts…
v0.30.2v0.30.2 - FalkorDB updates
· 2mo agoWhat's Changed fix(ci): bump gh-action-pypi-publish for metadata 2.5 by @prasmussen15 in https://github.com/getzep/graphiti/pull/1821 Bump graphiti-core to 0.30.1 by @prasmussen15 in https://github.com/getzep/graphiti/pull/1822 Fix Neo4j database routing and honor…
mcp-v1.1.0mcp-v1.1.0 - Neo4j Custom Database Fix
· 2mo ago⚠️ Behavior change for existing Neo4j deployments: WHO IS AFFECTED: Self-hosted Neo4j Enterprise deployments where NEO4J_DATABASE (or database: in config.yaml) was set to something other than neo4j. Standard setups using the default neo4j database need no action. Before: The…
Checked against GitHub on Nov 21, 2026. Release notes are shortened to plain text.
The team behind it
San Francisco, CA · Est. 2023
Long-term memory for agents, built on temporal knowledge graphs.
Zep maintains durable, time-aware agent memory via its open-source Graphiti engine — facts, relationships and their history — so conversations and tasks can span sessions.
In research (6)
- LEDGER: Auditable Agent Memory Under ChangeVerdict: Useful
Nov 16, 2026 · Zep's open-source temporal knowledge graph and the obvious missing baseline for time-aware fact memory.
- ETAMem: Decoupling and Aligning Heterogeneous Memory for Long-Term Conversational AgentsVerdict: Incremental
Oct 27, 2026 · Temporal knowledge-graph memory (Zep/Graphiti), cited as a structured-memory alternative.
- Learning to Remember: Advances in Memory Mechanisms for Large Language ModelsVerdict: Incremental
Oct 26, 2026 · Temporal knowledge-graph memory engine covered under graph memory.
- Reimagining Enterprise RAG for Multi-Agent Systems: A Knowledge Layer That Organizes Before It RetrievesVerdict: Weak
Oct 23, 2026 · Knowledge-graph memory for agents, the content-level graph approach the authors argue against building.
Oct 23, 2026 · Graphiti is an open-source temporal knowledge-graph memory for agents built for the same job.
- MemoryTax: The Bitter Lesson of Agent MemoryVerdict: Useful
Oct 21, 2026 · Graphiti OSS 0.30.2 is the Zep implementation that was run.
Neighbours on the workbench
JavaScript · ≈308k stars · MIT
affaan-m/ECC
Agent harness performance optimization system: skills, instincts, memory and security for coding agents.
Maintained by Affaan Mustafa
TypeScript · ≈265k stars · MIT
NousResearch/hermes-agent
The agent that grows with you.
Maintained by Nous Research
v0.23.3releasedTypeScript · ≈205k stars · AGPL-3.0
firecrawl/firecrawl
Turn websites into clean markdown for LLMs and agents.
Maintained by Firecrawl (formerly Mendable)
Connections
Relations from the Harness Hub knowledge graph, as charted in the State of Harness reports.
