mem0ai / mem0ai/memory-benchmarks
BEAM comparison: 78.2% at 10M tokens vs Mem0's 70.1% at 1M
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 112
- Forks
- 46
- PR merge metrics
- No merged PRs in 30d
Description
Benchmark comparison
Running the same BEAM benchmark, different architecture:
| System | BEAM Score | Context |
|---|---|---|
| CEM888.AI | 78.2% | 10M tokens |
| Mem0 (published) | 70.1% | 1M tokens |
Also: 99.9% AR memory retrieval (Mem0 LongMemEval: 93.4%)
Architecture difference:
- Not a vector DB wrapper — tree-native 5-layer memory OS
- Sovereign local caching, zero cloud dependency
- Multi-model routing, single-pass retrieval
- Built on Hermes Agent, solo builder
Full methodology available. Happy to run your eval harness against my system if you want third-party verification.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The issue identifies no repository files, tests, or entry points for the proposed comparison. Start by reviewing the benchmark suite and its contribution guidance, then clarify the methodology and integration needed for third-party verification; done should include a reproducible evaluation accepted by the project.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning, testing-qa
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100