Create Memory Config Node

C

AI/Memory

Create Memory Config

Creates a MemoryConfig that bundles database, embedding model, and tuning parameters for all memory nodes

memory_create_configllm

Inputs7

Outputs1

Security exposure2/10

Packagellm

Ratings

Scores range from 0 to 10. Higher values mean more impact, exposure, or operational weight.

SecurityAttack surface and exposure impact.

2/10High

PrivacyPotential sensitivity of processed data.

2/10High

PerformanceRuntime or resource pressure.

0/10High

GovernancePolicy, audit, or compliance impact.

2/10High

ReliabilityOperational stability considerations.

1/10High

CostExternal or compute cost impact.

0/10High

Input Pins

7

Database

Struct

database

LanceDB connection (from Open Database node). The table IS the scope boundary — use one table per user/session.

SchemaNodeDBConnectionNodeDBConnectionNodeDBConnection1 fields

cache_keystringrequired

Schema enforced

Embedding Model

Struct

embedding_model

Cached embedding model for vector search (from Load Embedding Model node)

SchemaCachedEmbeddingModelCachedEmbeddingModelCachedEmbeddingModel2 fields

cache_keystringrequired

model_typeBitTypesrequired

enum "Llm", "Vlm", "Tts", "Stt"...

Schema enforced

Max Context Tokens

Integer

max_context_tokens

Token budget for assembled memory context

Default 4000

Recall Strategy

String

recall_strategy

How to retrieve memories: recent_first (last N), relevance (vector similarity), hybrid (both)

Default hybrid

recent_firstrelevancehybrid

Recall Top K

Integer

recall_top_k

Max items returned from vector search

Default 20

Auto Compress

Boolean

auto_compress

Automatically compress old observations when threshold is reached

Default true

Compress Threshold

Integer

compress_threshold

Number of observations before triggering compression

Default 30

Output Pins

1

Memory Config

Struct

memory_config

Configured MemoryConfig — pass to any memory node

SchemaMemoryConfigMemoryConfigMemoryConfig7 fields

databaseNodeDBConnectionrequired

cache_keystringrequired

embedding_modelCachedEmbeddingModelrequired

cache_keystringrequired

model_typeBitTypesrequired

enum "Llm", "Vlm", "Tts", "Stt"...

max_context_tokensinteger:uint32required

format uint32min 0

recall_strategyRecallStrategyrequired

enum "RecentFirst", "Relevance", "Hybrid"

recall_top_kinteger:uint32required

format uint32min 0

auto_compressbooleanrequired

compress_thresholdinteger:uint32required

format uint32min 0

Schema enforced

Node Info

Internal name

memory_create_config

Category

AI/Memory

Version

1