API for managing personal memory items with authentication and user-specific data.
API for managing personal memory items with authentication and user-specific data.
This API supports three authentication methods:
X-API-Key headerX-API-Key: <your-api-key>X-Session-Token headerX-Session-Token: <your-session-token>Authorization headerAuthorization: Bearer <token>All endpoints require one of these authentication methods.
Text content for LLM metadata extraction
Base embedding vector (any dimensionality)
Domain for frequency schema selection (e.g. 'biomedical', 'code', 'general')
Domain for frequency schema selection (e.g. 'biomedical', 'code', 'general')
Domain for frequency schema selection (e.g. 'biomedical', 'code', 'general')
Specific frequency schema ID override (e.g. 'biomedical:scifact:2.0.0'). Takes precedence over domain.
Specific frequency schema ID override (e.g. 'biomedical:scifact:2.0.0'). Takes precedence over domain.
Specific frequency schema ID override (e.g. 'biomedical:scifact:2.0.0'). Takes precedence over domain.
Which output fields to return. Default: ['rotation_v3', 'metadata']. Request only what you need to minimize response size.
Which output fields to return. Default: ['rotation_v3', 'metadata']. Request only what you need to minimize response size.
Optional context metadata (createdAt, sourceType, customMetadata, etc.) to improve LLM extraction accuracy, especially for dates and entities.
Optional context metadata (createdAt, sourceType, customMetadata, etc.) to improve LLM extraction accuracy, especially for dates and entities.
curl -i -X POST \
https://memoryserver-staging-223473570766.us-west1.run.app/v1/holographic/transform \
-H 'Content-Type: application/json' \
-H 'X-API-Key: YOUR_API_KEY_HERE' \
-d '{
"content": "The patient presents with elevated troponin levels indicating myocardial damage",
"context_metadata": {
"createdAt": "2026-03-15T14:30:00Z",
"sourceType": "pubmed"
},
"domain": "biomedical",
"embedding": [
0.1,
-0.2,
0.3
],
"output": [
"rotation_v3",
"phases",
"metadata"
]
}'Successful Response
Inner data for transform response — only requested fields are populated.
Original embedding echoed back
Original embedding echoed back
Rotation V1 transform (same dims as input)
Rotation V1 transform (same dims as input)
Rotation V2 transform (same dims as input)
Rotation V2 transform (same dims as input)
Rotation V3 transform (recommended for search, same dims as input)
Rotation V3 transform (recommended for search, same dims as input)
Concatenation transform (input_dims + 196)
Concatenation transform (input_dims + 196)
14 raw phase values for on-device reconstruction and fast-path rerank
14 raw phase values for on-device reconstruction and fast-path rerank
LLM-extracted metadata keyed by frequency field name
LLM-extracted metadata keyed by frequency field name
{ "status": "success", "data": { "base_dim": 2560, "domain": "biomedical", "frequency_schema_id": "biomedical:scifact:2.0.0", "metadata": { … }, "phases": [ … ], "rotation_v3": [ … ], "timing_ms": 142.3 } }
Items to transform (max 50)
Domain for all items
Domain for all items
Domain for all items
Schema override for all items
Schema override for all items
Schema override for all items
curl -i -X POST \
https://memoryserver-staging-223473570766.us-west1.run.app/v1/holographic/transform/batch \
-H 'Content-Type: application/json' \
-H 'X-API-Key: YOUR_API_KEY_HERE' \
-d '{
"items": [
{
"id": "string",
"content": "string",
"embedding": [
0
],
"context_metadata": {}
}
],
"domain": "general",
"frequency_schema_id": "string",
"output": [
"base"
]
}'Successful Response
Inner data for transform response — only requested fields are populated.
Original embedding echoed back
Original embedding echoed back
Rotation V1 transform (same dims as input)
Rotation V1 transform (same dims as input)
Rotation V2 transform (same dims as input)
Rotation V2 transform (same dims as input)
Rotation V3 transform (recommended for search, same dims as input)
Rotation V3 transform (recommended for search, same dims as input)
Concatenation transform (input_dims + 196)
Concatenation transform (input_dims + 196)
14 raw phase values for on-device reconstruction and fast-path rerank
14 raw phase values for on-device reconstruction and fast-path rerank
LLM-extracted metadata keyed by frequency field name
LLM-extracted metadata keyed by frequency field name
Per-frequency metadata embeddings
Per-frequency metadata embeddings
Exact frequency schema ID used
{ "status": "success", "results": [ { … } ], "total": 0, "timing_ms": 0 }
The simplest entry point — zero infrastructure needed. Send your search query and candidate results, get back better-ranked results using CAESAR ensemble.
Auto-detection: candidates with phases use the fast path (~2-5ms each). Candidates with only content use the cold path (~100ms each, includes LLM extraction). You can mix both in a single request.
The search query text
Query embedding in the same space as candidate embeddings. If provided, used for cosine similarity. If omitted, computed server-side (Qwen 2560d).
Query embedding in the same space as candidate embeddings. If provided, used for cosine similarity. If omitted, computed server-side (Qwen 2560d).
Pre-computed query phases from a prior /transform call. If provided alongside query_embedding, skips LLM extraction entirely (hot path).
Pre-computed query phases from a prior /transform call. If provided alongside query_embedding, skips LLM extraction entirely (hot path).
Pre-computed query metadata embeddings from a prior /transform call (keyed by frequency string, e.g. '0.1'). Required for full HCond scoring with phase alignment.
Pre-computed query metadata embeddings from a prior /transform call (keyed by frequency string, e.g. '0.1'). Required for full HCond scoring with phase alignment.
Candidate documents to rerank (max 100)
Text content. Required for cold path (LLM extraction + cross-encoder).
Text content. Required for cold path (LLM extraction + cross-encoder).
Text content. Required for cold path (LLM extraction + cross-encoder).
Base embedding. If missing and content provided, computed server-side.
Base embedding. If missing and content provided, computed server-side.
Pre-computed phases from a prior /transform call. Enables fast path.
Pre-computed phases from a prior /transform call. Enables fast path.
Pre-computed SBERT metadata embeddings from a prior /transform call (keyed by frequency string, e.g. '0.1'). Enables full HCond scoring.
Pre-computed SBERT metadata embeddings from a prior /transform call (keyed by frequency string, e.g. '0.1'). Enables full HCond scoring.
Domain for frequency schema
Domain for frequency schema
Domain for frequency schema
Schema override
Schema override
Schema override
curl -i -X POST \
https://memoryserver-staging-223473570766.us-west1.run.app/v1/holographic/rerank \
-H 'Content-Type: application/json' \
-H 'X-API-Key: YOUR_API_KEY_HERE' \
-d '{
"candidates": [
{
"content": "Troponin is a cardiac biomarker released during myocardial injury...",
"id": "doc_1"
},
{
"content": "Aspirin reduces platelet aggregation...",
"id": "doc_2"
}
],
"domain": "biomedical",
"query": "How does troponin relate to myocardial infarction?",
"top_k": 10
}'Successful Response
Original retrieval score if provided
Original retrieval score if provided
Original retrieval score if provided
Per-method score breakdown (if return_scores=true)
Per-method score breakdown (if return_scores=true)
Present when cold path was used. Suggests storing phases for faster reranking.
Present when cold path was used. Suggests storing phases for faster reranking.
Present when cold path was used. Suggests storing phases for faster reranking.
{ "status": "success", "data": { "rankings": [ … ], "ensemble_used": "string", "domain": "string", "timing_ms": 0, "optimization_hint": "string" } }
Domain for frequency schema
Domain for frequency schema
Domain for frequency schema
Schema override
Schema override
Schema override
curl -i -X POST \
https://memoryserver-staging-223473570766.us-west1.run.app/v1/holographic/metadata \
-H 'Content-Type: application/json' \
-H 'X-API-Key: YOUR_API_KEY_HERE' \
-d '{
"content": "string",
"domain": "general",
"frequency_schema_id": "string",
"context_metadata": {}
}'{ "status": "success", "data": { "metadata": {}, "phases": [ … ], "domain": "string", "frequency_schema_id": "string", "timing_ms": 0 } }
curl -i -X GET \
https://memoryserver-staging-223473570766.us-west1.run.app/v1/holographic/domains \
-H 'X-API-Key: YOUR_API_KEY_HERE'{ "status": "success", "domains": [ { … } ], "total": 0, "shortcuts": { "property1": "string", "property2": "string" } }
Schema name in format 'company:domain:version' (e.g. 'acme:support_tickets:1.0.0')
Human-readable description
Human-readable description
Human-readable description
Frequency field definitions (1-14 fields, one per frequency band)
Hz value (must be one of the 14 standard frequencies: 0.1, 0.5, 2.0, 4.0, 6.0, 10.0, 12.0, 18.0, 19.0, 24.0, 30.0, 40.0, 50.0, 70.0)
Field type
Allowed values for enum type
Allowed values for enum type
Field description
Field description
Field description
curl -i -X POST \
https://memoryserver-staging-223473570766.us-west1.run.app/v1/holographic/domains \
-H 'Content-Type: application/json' \
-H 'X-API-Key: YOUR_API_KEY_HERE' \
-d '{
"description": "Support ticket classification schema",
"fields": [
{
"frequency": 4,
"name": "priority",
"type": "enum",
"values": [
"P0",
"P1",
"P2",
"P3"
],
"weight": 0.9
},
{
"frequency": 6,
"name": "component",
"type": "free_text",
"weight": 0.7
},
{
"frequency": 12,
"name": "resolution_type",
"type": "enum",
"values": [
"bug_fix",
"config",
"wontfix"
],
"weight": 0.8
}
],
"name": "acme:support_tickets:1.0.0"
}'{ "status": "success", "schema_id": "string", "domain": "string", "num_frequencies": 0 }