Memory + ModelsMemory + Models
Memory + Models
Overview of Coven memory layers, memory search, model access, provider boundaries, and current OpenAI/Anthropic/local model posture.
1 min read
Memory is what turns a one-off harness session into a persistent familiar. Models are the provider side of that familiar's reasoning, but Coven does not become a model gateway. The durable runtime boundary is: Coven records sessions and events; clients/familiars derive memory from those events; harnesses own provider auth and model access.
Rendering diagram…
Section map
- Memory overview defines the four memory layers.
- Working memory covers in-session context.
- Persistent memory covers durable facts and identity.
- Episodic memory covers specific sessions, decisions, and outcomes.
- Semantic memory covers embeddings and meaning-based recall.
- Memory search covers keyword and semantic lookup.
- Models explains why Coven is not a model gateway.
- Provider boundary names the credential boundary.
- OpenAI, Anthropic, and Local models cover provider-specific paths.
Related
Was this page helpful?No
Last updated on