Preview. These maps are our reading of public sources and haven't been checked with the projects' maintainers yet.
MemoryOS (BAI-LAB)
Research memory operating system (EMNLP 2025 oral) with short-, mid- and long-term persona memory tiers, plus a pip package and MCP server.
github.com/BAI-LAB/MemoryOSMaintenance
Last release V1.2 on 2025-07-18; occasional fixes merged since (latest 2026-07-07). (checked 2026-10-25)
In Agent memory
Where it stands apart
- Is it actively maintained? maintenance or slowing (5% of 21 others)
- How is it mainly sustained? community, foundation, or research (5% of 20 others)
- What is the main storage model for long-term memory? structured records: profiles, tiers, or tables (15% of 20 others)
What it shares with most of the community
- How is the main thing licensed? permissive open source (MIT/Apache/BSD) (67% of 21 others)
- Can it run entirely on your own device or infrastructure, without the maker's servers? yes (67% of 21 others)
- Who is it for? developers (76% of 21 others)
- In the default setup, who mainly decides what gets remembered? the system, by extracting memories from conversations or data in the background (67% of 21 others)
- How can the person the memories are about see and edit them? only through developer APIs or a developer dashboard (62% of 21 others)
- By default, how do old or wrong memories mainly go away? the system updates, merges, or decays them automatically (58% of 19 others)
- Does it work with models from different LLM vendors? yes, model-agnostic (86% of 21 others)
- Is there a first-party MCP server for it? yes (67% of 21 others)
- How is memory separated between users or tenants? built-in scoping by user, agent, session, or tenant IDs (62% of 21 others)
- Does it publish scores on public memory benchmarks (LoCoMo, LongMemEval, BEAM, etc.)? yes (62% of 21 others)
Nearest neighbors
- A-MEM 71% of 17 shared answers match
- Graphiti 71% of 17 shared answers match
- LangMem 71% of 17 shared answers match
Every answer
| Question | Answer | Source |
|---|---|---|
| Is it actively maintained? | stands apart maintenance or slowing | source 2026-10-25 |
| How is the main thing licensed? | permissive open source (MIT/Apache/BSD) Apache-2.0 | source 2026-10-25 |
| How is it mainly sustained? | stands apart community, foundation, or research From BaiJia AI research team at BUPT (EMNLP 2025); no pricing, hosted playground is invite-only. | source 2026-10-26 |
| Can it run entirely on your own device or infrastructure, without the maker's servers? | yes | source 2026-10-25 |
| Do you need an account to use it? | no account Open-source code via PyPI/Docker needs no maker account; only your own LLM key. | source 2026-10-26 |
| Who is it for? | developers | source 2026-10-25 |
| Which shared standard or format does it mainly build on? | none of these (its own format or interface) Own tiered store; MCP one interface. | source 2026-10-26 |
| In the default setup, who mainly decides what gets remembered? | the system, by extracting memories from conversations or data in the background Automated user profile and knowledge updating. | source 2026-10-25 |
| How can the person the memories are about see and edit them? | only through developer APIs or a developer dashboard | source 2026-10-25 |
| By default, how do old or wrong memories mainly go away? | the system updates, merges, or decays them automatically Heat-based promotion and eviction between tiers; consolidation from short to mid term. | source 2026-10-25 |
| Where do memories live in the default setup the docs lead with? | in the developer's own infrastructure | source 2026-10-25 |
| Does it work with models from different LLM vendors? | yes, model-agnostic OpenAI-compatible APIs and local embeddings. | source 2026-10-25 |
| What is the main storage model for long-term memory? | stands apart structured records: profiles, tiers, or tables Hierarchical short/mid/long-term tiers (ChromaDB optional). | source 2026-10-25 |
| What shape does it mainly take (the one its docs lead with)? | library or API primitive you build into your app | source 2026-10-25 |
| Is there a first-party MCP server for it? | yes | source 2026-10-25 |
| How is memory separated between users or tenants? | built-in scoping by user, agent, session, or tenant IDs user_id / assistant_id parameters. | source 2026-10-25 |
| Does it publish scores on public memory benchmarks (LoCoMo, LongMemEval, BEAM, etc.)? | yes LoCoMo results in paper. | source 2026-10-25 |
Sources
This page was derived from:
- GitHub · github.com/BAI-LAB/MemoryOS · read 2026-10-25
- Web page · arxiv.org/abs/2506.06326 · read 2026-10-25
RSS for this project · Its own release feed
Are you a maintainer? Claiming and correcting projects is coming soon.