open-brainy

v7.3.0 1a813c7b86

v7.3.0 - Progressive Init & Readiness API Stable

dpsifr released this 2026-01-07 21:52:12 +01:00 | 725 commits to main since this release

Full Changelog: https://github.com/soulcraftlabs/brainy/compare/v7.2.2...v7.3.0

Downloads

Source code (ZIP)
2 downloads
Source code (TAR.GZ)
4 downloads

v7.2.2 19c5e67035

v7.2.2 - Performance Optimizations Stable

dpsifr released this 2026-01-07 19:46:59 +01:00 | 727 commits to main since this release

Full Changelog: https://github.com/soulcraftlabs/brainy/compare/v7.2.1...v7.2.2

Downloads

Source code (ZIP)
3 downloads
Source code (TAR.GZ)
2 downloads

v7.2.1 cf06d82520

v7.2.1 - bun --compile model loading fix Stable

dpsifr released this 2026-01-07 03:01:38 +01:00 | 730 commits to main since this release

Full Changelog: https://github.com/soulcraftlabs/brainy/compare/v7.2.0...v7.2.1

Downloads

Source code (ZIP)
2 downloads
Source code (TAR.GZ)
2 downloads

v7.2.0 1a32ed65ab

v7.2.0 - 580x Faster Embedding Initialization Stable

dpsifr released this 2026-01-07 02:19:58 +01:00 | 732 commits to main since this release

Full Changelog: https://github.com/soulcraftlabs/brainy/compare/v7.1.1...v7.2.0

Downloads

Source code (ZIP)
2 downloads
Source code (TAR.GZ)
5 downloads

v7.1.1 4fc1fb7340

v7.1.1 - Critical Performance Fix for Cloud Storage Stable

dpsifr released this 2026-01-07 01:03:55 +01:00 | 734 commits to main since this release

Critical Bug Fix

Fixed 50-100x slower add() operations on cloud storage (GCS/S3/R2/Azure)

Symptoms

add() taking 7-12 seconds instead of 50-200ms
Only affects cloud storage with auto-detection (not explicit type: 'gcs')

Root Cause

Storage type detection in setupIndex() relied on this.config.storage.type which was never set after createStorage() auto-detected the storage type. This caused cloud storage to use 'immediate' persistence mode instead of 'deferred', resulting in 20-30 GCS writes per add() operation.

Fix

Added getStorageType() helper that detects storage type from the storage instance class name (e.g., GcsStorage → 'gcs'), used as fallback when config.storage.type is not explicitly set.

Upgrade Instructions

Simply update to v7.1.1:

npm install @soulcraft/brainy@7.1.1

No configuration changes required - the fix automatically detects cloud storage types.

Workaround for v7.1.0 users (if you can't upgrade immediately)

const brain = new Brainy({
  storage: {
    type: 'gcs',  // Explicit type fixes the issue
    gcsNativeStorage: { bucketName: 'your-bucket' }
  },
  hnswPersistMode: 'deferred'  // Or explicitly set this
})

Performance Tests Added

Added regression tests to prevent future issues:

Single add() < 500ms
10 add() operations < 5 seconds
Storage type detection verification for GCS/S3/R2/Azure

Full Changelog: https://github.com/soulcraftlabs/brainy/compare/v7.1.0...v7.1.1

Downloads

Source code (ZIP)
1 download
Source code (TAR.GZ)
1 download

v7.1.0 61100bdc9d

v7.1.0 - New Embedding & Analysis APIs Stable

dpsifr released this 2026-01-06 23:53:13 +01:00 | 737 commits to main since this release

New APIs

6 New Public APIs leveraging the Candle WASM embedding engine and optimized indexes:

API	Description	Performance
`embedBatch(texts)`	Batch embed multiple texts	Batch WASM processing - avoids N separate JS↔WASM calls
`similarity(textA, textB)`	Semantic similarity score (0-1)	Single call vs manual embed + embed + cosine
`indexStats()`	Comprehensive index statistics	O(1) - aggregates pre-computed stats
`neighbors(entityId, options)`	Graph traversal with filters	O(log n) - LSM-tree with bloom filters, sub-5ms
`findDuplicates(options)`	Find semantic duplicates	O(k log n) - uses HNSW for ANN search
`cluster(options)`	Cluster by similarity	O(k log n) - greedy algorithm with HNSW

Performance Stack (v7.0.0+)

The new APIs leverage the optimized infrastructure introduced in v7.0.0:

Component	Technology	Benefit
Embeddings	Candle WASM (Rust)	93MB binary with embedded MiniLM-L6-v2, zero downloads
Vector Search	HNSW Index	O(log n) approximate nearest neighbor
Graph Traversal	LSM-tree + Bloom Filters	90% of queries skip disk I/O, sub-5ms lookups
Metadata Filtering	RoaringBitmap32	Compressed bitmaps for fast AND/OR operations

Examples

// Batch embedding
const embeddings = await brain.embedBatch([
  'Machine learning',
  'Deep neural networks',
  'Natural language processing'
])

// Semantic similarity
const score = await brain.similarity(
  'The cat sat on the mat',
  'A feline was resting on the rug'
) // ~0.85

// Find duplicates
const duplicates = await brain.findDuplicates({
  threshold: 0.9,
  type: NounType.Document
})

// Clustering
const clusters = await brain.cluster({
  threshold: 0.8,
  minClusterSize: 3,
  includeCentroid: true
})

// Graph neighbors
const neighbors = await brain.neighbors(entityId, {
  direction: 'outgoing',
  depth: 2
})

// Index statistics
const stats = await brain.indexStats()
console.log(`${stats.entities} entities, ${stats.vectors} vectors`)

Migration from v6.x

v7.0.0 introduced breaking changes to the embedding system:

Removed: onnxruntime-node dependency (was 200MB+ with external model downloads)
Added: Candle WASM with embedded model weights (93MB, zero-config)
Removed: Semantic type inference (NLP-based type detection)
Works in: Node.js, Bun, Bun --compile, browsers

Full Changelog: https://github.com/soulcraftlabs/brainy/compare/v7.0.1...v7.1.0

Downloads