BrainAPI is a knowledge graph–powered AI memory layer that transforms unstructured data into structured knowledge, enabling intelligent search, recommendations, and contextual memory for AI agents and applications.
Python ai-agentsai-memoryknowledge-graphknowledge-managementknowledge-representationreccomendation-system
Commits on main
Add recommend and synergy retrieval with recsys benchmarks.
1e4cf88 Christian committed 3 months ago
Add BEAM 10M harness: plan concat normalize and campaign scripts.
cf61c99 Christian committed 3 months ago
fix: retry wrapped Hugging Face 429s during Docker model preload
2fe306f Christian committed 3 months ago
docs(readme): fix broken public docs links and context notes
6a1ec7f Cursor Agent committed 3 months ago
fix(retrieve): default preferred_entities to empty string
ac8e8af Cursor Agent committed 3 months ago
chore(benchmarks): force-track key BEAM and LoCoMo run artifacts
4b97204 Christian committed 3 months ago
chore: ignore Redis dump.rdb and shrink Docker build context
cdcd002 Christian committed 3 months ago
chore(benchmarks): update REPORTS ledger and multi-suite docs
0b6f31c Christian committed 3 months ago
docs(research): add BEAM/LongMemEval protocols and efficiency plans
1025e06 Christian committed 3 months ago
feat(benchmarks): harden LoCoMo SOTA answer and reporting path
1cc698d Christian committed 3 months ago
feat(retrieval): support history-aware retrieve and graph store helpers
35e3479 Christian committed 3 months ago
feat(ingest): batch Architect extract with scratchpad, grounding, and cost tracking
dd4e9e3 Christian committed 3 months ago
feat(benchmarks): SOTA LoCoMo harness and multi-suite REPORTS ledger
785a507 Christian committed 3 months ago
feat(retrieval): Event GraphRAG hubs, topics, and provenance
664027b Christian committed 3 months ago
test(ingestion+retrieval): test improvements after locomo benchmark
82f41d2 Christian committed 3 months ago