Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.gridcertainty.com/llms.txt

Use this file to discover all available pages before exploring further.

All MemWire behaviour is controlled through MemWireConfig. Choose your vector store, embedding model, and LLM provider, then tune recall and graph settings to fit your use case. 
from memwire import MemWire, MemWireConfig

config = MemWireConfig(
    qdrant_path="./memwire_data",
    qdrant_collection_prefix="app_",
)
memory = MemWire(config=config)

​
Components

Vector Databases

Connect to a vector store.

Embedding Models

Choose embedding model.

LLMs

Use any LLM provider with recalled context.

SQL Databases

Choose where to store metadata ledger for memories, graph edges, and categories.

​
Recall tuning

Recall assembles relevant memories into a formatted context string for injection into LLM prompts. 
config = MemWireConfig(
    recall_min_relevance=0.25,   # minimum similarity score to include a memory
    recall_max_paths=10,         # max memory paths returned
    recall_max_depth=4,          # BFS depth when traversing the memory graph
    recall_seed_top_k=5,         # top-k seed nodes before graph traversal
    tension_threshold=0.6,       # similarity above which two memories are flagged as contradictory
    recency_weight=0.3,          # weight of recency vs relevance in scoring
    recency_halflife=3600.0,     # half-life in seconds for recency decay
)
ParameterDefaultDescription
recall_min_relevance0.25Minimum cosine similarity for a memory to be included in recall.
recall_max_paths10Maximum number of memory paths returned per recall call.
recall_max_depth4Maximum BFS depth when traversing the memory graph.
recall_seed_top_k5Number of top-k seed nodes to start graph traversal from.
recall_bfs_max_branch5Maximum branches per node during BFS traversal.
recall_bfs_max_paths200Maximum candidate paths evaluated before pruning.
tension_threshold0.6Similarity above which two memories are flagged as contradictory.
recency_weight0.3Weight given to recency vs semantic relevance when scoring memories.
recency_halflife3600.0Half-life in seconds for recency decay. Lower = faster decay.

​
Graph construction

MemWire organises memories in a graph. Edges encode semantic relationships between memories. 
config = MemWireConfig(
    displacement_threshold=0.15,
    node_merge_similarity=0.85,
    edge_weight_default=0.5,
    edge_decay_rate=0.02,
    edge_reinforce_amount=0.1,
)
ParameterDefaultDescription
displacement_threshold0.15Minimum similarity required to create an edge between two memories.
node_merge_similarity0.85Similarity above which two memory nodes are merged (deduplication).
edge_weight_default0.5Initial weight for new edges.
edge_weight_min0.01Floor for edge weights after decay.
edge_weight_max1.0Ceiling for edge weights after reinforcement.
edge_decay_rate0.02Rate at which unused edges decay per cycle.
edge_reinforce_amount0.1Amount added to an edge weight on positive feedback.
cross_memory_recent_limit50Number of recent memories considered when building cross-memory edges.

​
Memory classification

Memories are automatically classified into categories using zero-shot cosine similarity.
CategoryDescription
factFactual statements or pieces of information
preferencePersonal preferences or opinions
instructionDirectives or rules to follow
eventThings that happened
entityInformation about a person, place, or organisation
config = MemWireConfig(
    classification_threshold=0.05,
    default_anchors={
        "fact": ["This is a factual statement"],
        "preference": ["This is a personal preference"],
        "instruction": ["This is a rule or directive to follow"],
        "event": ["This is something that happened"],
        "entity": ["This is about a specific person or organisation"],
    }
)

​
Feedback loop

config = MemWireConfig(
    feedback_strengthen_rate=0.1,
    feedback_weaken_rate=0.05,
    feedback_align_strengthen=0.5,
    feedback_align_weaken=0.2,
)
ParameterDefaultDescription
feedback_strengthen_rate0.1Amount to strengthen edges that contributed to a good response.
feedback_weaken_rate0.05Amount to weaken edges that did not contribute.
feedback_align_strengthen0.5Alignment score threshold above which edges are strengthened.
feedback_align_weaken0.2Alignment score threshold below which edges are weakened.

​
SQL Databases

MemWire stores memory metadata — content, graph edges, categories, and access counts — in a SQL database, while all embedding vectors live separately in a vector store. SQLite is the default because it requires no server or configuration, letting you run MemWire locally out of the box; switch to PostgreSQL or any other SQLAlchemy-compatible database by setting database_url. 
config = MemWireConfig(
    database_url="sqlite:///memwire.db",
    org_id="default",
)
ParameterDefaultDescription
database_urlNoneSQLAlchemy-compatible URL. Defaults to sqlite:///memwire_{org_id}.db.
org_id"default"Organisation identifier for multi-tenant isolation.

​
Performance

config = MemWireConfig(
    background_threads=2,
    embedding_cache_maxsize=10000,
)
ParameterDefaultDescription
background_threads2Background threads for async graph and storage writes.
embedding_cache_maxsize10000LRU cache size for embedding vectors.
from memwire import MemWire, MemWireConfig

config = MemWireConfig(
    qdrant_path="./memwire_data",
    qdrant_collection_prefix="app_",
)
memory = MemWire(config=config)
No environment variables are required — every setting is an explicit Python argument with a sensible default.

​
Search quality

Control how memories are retrieved during search and recall. 
config = MemWireConfig(
    use_hybrid_search=True,    # combine dense + sparse vectors
    use_reranking=False,       # apply a cross-encoder reranker
    reranker_model_name="Xenova/ms-marco-MiniLM-L-6-v2",
)
ParameterDefaultDescription
use_hybrid_searchTrueCombine dense and sparse vectors for more accurate retrieval.
use_rerankingFalseApply a cross-encoder reranker to re-score top candidates.
reranker_model_nameXenova/ms-marco-MiniLM-L-6-v2Reranker model (used when use_reranking=True).

​
Recall tuning

Recall assembles relevant memories into a formatted context string for injection into LLM prompts. 
config = MemWireConfig(
    recall_min_relevance=0.25,   # minimum similarity score to include a memory
    recall_max_paths=10,         # max memory paths returned
    recall_max_depth=4,          # BFS depth when traversing the memory graph
    recall_seed_top_k=5,         # top-k seed nodes before graph traversal
    tension_threshold=0.6,       # similarity above which two memories are flagged as contradictory
    recency_weight=0.3,          # weight of recency vs relevance in scoring
    recency_halflife=3600.0,     # half-life in seconds for recency decay
)
ParameterDefaultDescription
recall_min_relevance0.25Minimum cosine similarity for a memory to be included in recall.
recall_max_paths10Maximum number of memory paths returned per recall call.
recall_max_depth4Maximum BFS depth when traversing the memory graph.
recall_seed_top_k5Number of top-k seed nodes to start graph traversal from.
recall_bfs_max_branch5Maximum branches per node during BFS traversal.
recall_bfs_max_paths200Maximum candidate paths evaluated before pruning.
tension_threshold0.6Similarity above which two memories are flagged as contradictory.
recency_weight0.3Weight given to recency vs semantic relevance when scoring memories.
recency_halflife3600.0Half-life in seconds for recency decay. Lower = faster decay.

​
Graph construction

MemWire organises memories in a graph. Edges encode semantic relationships between memories. 
config = MemWireConfig(
    displacement_threshold=0.15,  # min similarity to create an edge
    node_merge_similarity=0.85,   # similarity above which two nodes are merged (dedup)
    edge_weight_default=0.5,
    edge_weight_min=0.01,
    edge_weight_max=1.0,
    edge_decay_rate=0.02,         # edges decay over time when not reinforced
    edge_reinforce_amount=0.1,    # amount edges are boosted on positive feedback
)
ParameterDefaultDescription
displacement_threshold0.15Minimum similarity required to create an edge between two memories.
node_merge_similarity0.85Similarity above which two memory nodes are merged (deduplication).
edge_weight_default0.5Initial weight for new edges.
edge_weight_min0.01Floor for edge weights after decay.
edge_weight_max1.0Ceiling for edge weights after reinforcement.
edge_decay_rate0.02Rate at which unused edges decay per cycle.
edge_reinforce_amount0.1Amount added to an edge weight on positive feedback.
cross_memory_recent_limit50Number of recent memories considered when building cross-memory edges.

​
Memory classification

Memories are automatically classified into categories. Classification uses zero-shot cosine similarity against anchor phrases.
CategoryDescription
factFactual statements or pieces of information
preferencePersonal preferences or opinions
instructionDirectives or rules to follow
eventThings that happened
entityInformation about a person, place, or organisation
config = MemWireConfig(
    classification_threshold=0.05,  # minimum score margin between top categories
)
You can override the anchor phrases for each category: 
config = MemWireConfig(
    default_anchors={
        "fact": ["This is a factual statement"],
        "preference": ["This is a personal preference"],
        "instruction": ["This is a rule or directive to follow"],
        "event": ["This is something that happened"],
        "entity": ["This is about a specific person or organisation"],
    }
)

​
Feedback loop

The feedback loop reinforces graph edges that led to good responses, and weakens edges that did not contribute. 
config = MemWireConfig(
    feedback_strengthen_rate=0.1,    # edge boost on positive feedback
    feedback_weaken_rate=0.05,       # edge reduction on negative feedback
    feedback_align_strengthen=0.5,
    feedback_align_weaken=0.2,
)
ParameterDefaultDescription
feedback_strengthen_rate0.1Amount to strengthen edges that contributed to a good response.
feedback_weaken_rate0.05Amount to weaken edges that did not contribute.
feedback_align_strengthen0.5Alignment score threshold above which edges are strengthened.
feedback_align_weaken0.2Alignment score threshold below which edges are weakened.
Call memory.feedback() after each LLM response to close the loop: 
result = memory.recall("How should I format my answers?", user_id="alice")
# ... call your LLM with result.formatted as context ...
memory.feedback(response="<llm response here>", user_id="alice")

​
Storage

config = MemWireConfig(
    database_url="sqlite:///memwire.db",  # SQLAlchemy-compatible URL
    org_id="default",                     # organisation identifier for multi-tenant isolation
)
ParameterDefaultDescription
database_urlNoneSQLAlchemy-compatible database URL. Defaults to sqlite:///memwire_{org_id}.db.
org_id"default"Organisation identifier. Used as a namespace for multi-tenant deployments.

​
Performance

config = MemWireConfig(
    background_threads=2,         # threads for async background writes
    embedding_cache_maxsize=10000, # LRU cache for embedding vectors
)
ParameterDefaultDescription
background_threads2Number of background threads for async graph and storage writes.
embedding_cache_maxsize10000LRU cache size for embedding vectors. Increase for large workloads.