Quantum Graph
Memory
QGM · Time-aware graph memory
Time-aware graph memory for QAG agents — provenance and temporal scope preserved across sessions and regulatory replays.
Quantum Graph Memory (QGM) — request a demo
Target: later in 2026
Request a demo — early access for teams with a 'replay-as-of' requirement.
"Replay as of March 1" is a memory question, not a retrieval question
Classical memory answers "what is closest now." QGM answers "what was true then."
Classical memory
Flat store. Present tense.
- Vector similarity against "now" — no explicit time axis
- Provenance as metadata — losable, stampable, brittle
- No cross-fact reasoning — dependencies implicit at best
QGM
Graph. Time-aware. Auditable.
- Validity intervals on every fact — temporal queries are first-class
- Provenance is a graph edge, not a metadata tag
- Cross-fact dependencies reasoned by the QAG Engine directly
The memory layer, regulatory-grade
Time-aware by construction
Every fact and rule carries a validity interval. Temporal queries are first-class.
Provenance preserved end-to-end
Source, author, and revision chain on every memory item — replayable years later.
Graph, not key-value
Memory is a graph over Q-Prime encodings. Cross-fact dependencies are first-class.
Cross-session continuity
Agents retain consistent memory across sessions — no summarize-and-forget.
Regulatory replay
Answers from memory as-of a given date — rule versions and conflicts in force then.
QAG-native
Speaks the same seven HSC signals as the QAG Engine. Audit trail is built in.
Questions regulated teams ask about graph memory.
What is Quantum Graph Memory (QGM)?
How is QGM different from a vector or graph database?
What is regulatory replay?
When will QGM be available?
Need "replay as of" in your workflow?
Request a demo and tell us your regulatory replay requirement.