Goal: make the memory benchmark compare WaveMind with agent-memory systems, not only vector stores.
Targets:
- Mem0 baseline
- Zep baseline
- LangGraph persistent memory baseline
Expected output:
- adapters under
benchmarks/ or wavemind/integrations/ where appropriate;
- one reproducible benchmark command;
- JSON result artifact under
benchmarks/;
- README summary with precision/recall/latency and a short interpretation.
Important: this should stay honest. If a dependency needs an API key, cloud service, or paid setup, document that clearly and keep the offline path runnable.
Goal: make the memory benchmark compare WaveMind with agent-memory systems, not only vector stores.
Targets:
Expected output:
benchmarks/orwavemind/integrations/where appropriate;benchmarks/;Important: this should stay honest. If a dependency needs an API key, cloud service, or paid setup, document that clearly and keep the offline path runnable.