v2.1.0 — SOTA Gap Implementations

v2.1.0 — State-of-the-Art Gap Implementations

13 new modules across 3 crates addressing gaps between RuVector and 2024-2026 research from Google, Meta, DeepSeek, and Microsoft. 8,577 lines of new code, 859 tests passing, zero regressions.

Highlights

Advanced Search & Retrieval (ruvector-core)

• Hybrid Search (RRF) — Sparse + dense vector fusion with Reciprocal Rank Fusion, SPLADE-compatible scoring. 20-49% retrieval improvement.
• Graph RAG — Knowledge graph + Leiden community detection + local/global/hybrid search. 30-60% improvement on complex multi-hop queries.
• DiskANN / Vamana — SSD-backed billion-scale ANN with alpha-RNG pruning and LRU page cache. <10ms latency.
• ColBERT Multi-Vector — Per-token late interaction retrieval with MaxSim, AvgSim, SumMax scoring.
• Matryoshka Embeddings — Adaptive-dimension search with funnel and cascade modes for speed with minimal recall loss.
• OPQ — Optimized Product Quantization with learned rotation matrix. 10-30% error reduction vs standard PQ.
• LSM Compaction — Log-Structured Merge-tree for write-heavy workloads with bloom filters.

Attention & Inference (ruvector-attention)

• FlashAttention-3 — IO-aware tiled attention reducing memory from O(N²) to O(N). Configurable block sizes, causal masking, dropout.
• Multi-Head Latent Attention (MLA) — DeepSeek-V2/V3 style KV-cache compression (~93% reduction).
• KV-Cache Compression — 3-4 bit asymmetric per-channel quantization (TurboQuant-inspired). H2O, Sliding Window, PyramidKV eviction. 6-8x memory reduction.
• Selective State Space Models (Mamba) — Linear-time sequence processing with selective scan and discretization.
• Speculative Decoding — Draft-verify pipeline with Medusa multi-head and tree attention for 2-3x generation speedup.

Graph Learning (ruvector-gnn)

• GraphMAE — Graph Masked Autoencoder with GAT encoder, SCE loss, degree-centrality masking, re-masking regularization.

Quality

• 859 Rust tests — 423 (core) + 210 (attention) + 226 (gnn), all passing
• Zero regressions from v2.0.6
• No unsafe code in any new module
• Security fixes: NaN-safe sort comparisons, quantization input validation

Published Packages

crates.io:

Crate	Version
ruvector-core	2.1.0
ruvector-attention	2.1.0
ruvector-gnn	2.1.0
ruvector-attention-wasm	2.1.0
ruvector-gnn-wasm	2.1.0
ruvllm	2.1.0

npm:

Package	Version
ruvector	0.2.19
ruvector-wasm	2.1.0
ruvector-attention-wasm	2.1.0
ruvector-gnn-wasm	2.1.0
ruvector-attention-unified-wasm	0.1.0
@ruvector/ruvllm	2.5.4

CLI Fix

• Fixed `npx ruvector create` and `benchmark` commands — `dimension` → `dimensions` field name mismatch (#307)

Documentation

• Updated root README with all new SOTA modules
• Updated npm README with v2.1 features and TurboQuant section
• Updated @ruvector/ruvllm README with TurboQuant KV-cache compression docs
• ADR-128: SOTA gap analysis and implementation documentation

Full Changelog: v2.0.6...v2.1.0

---

Training Pipeline (ADR-129)

Added complete GCloud training infrastructure for continuous model improvement:

• Release gate automation — 7 ship/no-ship criteria (G1-G7) with automated checker
• Dataset governance — Schema validation, dedup, contamination checks, quality scoring
• Nightly training — Incremental LoRA from pi.ruv.io brain learnings → validate → push to HF
• TurboQuant sidecar — .turboquant.json per-layer KV-cache config profiles
• Cloud Run Jobs — 4 GPU jobs (calibration, SFT, benchmark, nightly) + 2 schedulers
• Ablation matrix — 5-run isolation testing (baseline → imatrix → SFT → DPO → TQ)

Deploy: ./scripts/training/deploy_training.sh

v0.5.0-dragnes: DrAgnes + Common Crawl + Gemini Grounding

DrAgnes Dermatology AI + Common Crawl WET Pipeline + Gemini Grounding Agents

New Features

• DrAgnes (examples/dragnes/): Standalone AI dermatology intelligence platform with CNN classification, HIPAA compliance, DermLite integration
• Common Crawl WET Pipeline: 178-domain automated import (medical + CS + physics)
• Gemini Grounding Agents: 4 autonomous agents — fact verification, relation generation, cross-domain discovery, research director
• Brain: 2,064 memories, 943K edges, 57x sparsifier, Gemini 2.5 Flash optimizer

Bug Fixes

• SONA getStats().trajectoriesRecorded always 0 (#273)
• SONA state persistence across restarts (#274)
• MCP SSE auto-reconnect on stale sessions
• RuVocal icon 404, FoundationBackground crash

Published Packages

• ruvector-sona@0.1.7 (crates.io)
• ruvector@0.2.17 (npm)

ADRs

117 (DrAgnes), 118 (Cost Strategy), 119 (Historical Crawl), 120 (WET Pipeline), 121 (Gemini Grounding), 122 (Grounding Agents)

Infrastructure

• Cloud Run: 4 CPU / 4 GiB / 20 instances
• 19 scheduler jobs (train, drift, transfer, graph, WET, Gemini agents)
• 3 Cloud Run Jobs (WET import, Gemini agents, rvagent learning)

v0.88.0 — RuVix Cognition Kernel + Neural Trader WASM

Highlights

RuVix Cognition Kernel (ADR-087)

A foundational architecture for RuVix — a cognition kernel for the Agentic Age. Not a Linux clone, but a purpose-built OS with 6 primitives, 12 syscalls, and RVF as native boot object.

Core primitives: task, capability, region, queue, timer, proof

Key innovations:

• Proof-gated mutation as kernel invariant — no proof, no mutation
• RuVector kernel-resident — vectors and graphs are native kernel resources
• Coherence-aware scheduler — deadline + novelty + structural risk
• RVF as boot object — signed packages are complete cognitive units
• 12 syscalls total

Build path: Phase A (Linux-hosted nucleus, 9 crates) → Phase B (bare metal AArch64)

8 demo applications: proof-gated vector journal, edge ML inference, drone swarm, self-healing knowledge graph, collective intelligence mesh, quantum-coherent memory replay, biological signal processor, adversarial reasoning arena

Security Hardening (Post-Audit)

6 specification clarifications added after security audit:

1. Root task privilege attenuation — drops capabilities after boot
2. Capability delegation depth limit — max depth 8, GRANT_ONCE right
3. Boot RVF proof bootstrap — signed boot is the single trusted path
4. Reflex proof cache scoping — per-(mutation_hash, nonce), single-use
5. Zero-copy IPC TOCTOU — rejects descriptors into Slab regions
6. Boot signature failure — kernel panic, no fallback

Neural Trader WASM Bindings (ADR-086)

4 new crates: core (2 tests), coherence (7 tests), replay (3 tests), wasm (10 + 43 JS tests)

Visual Updates

Animated dot-matrix grid and billboard squares on pi.ruv.io

PRs: #244, #248 + security hardening commit

🤖 Generated with claude-flow

Release v2.0.5

RuVector Release v2.0.5

Published Packages

crates.io

• ruvector-math - Advanced math primitives
• ruvector-attention - 7-theory attention mechanisms
• ruvector-math-wasm - WASM bindings for math
• ruvector-attention-wasm - WASM bindings for attention

npm

• @ruvector/math-wasm - Browser WASM package
• @ruvector/attention - Main Node.js package (auto-selects platform)
• @ruvector/attention-wasm - Browser WASM package
• Platform-specific: linux-x64, linux-arm64, darwin-x64, darwin-arm64, win32-x64

Installation

# Rust
cargo add ruvector-math ruvector-attention

# Node.js (auto-selects correct binary)
npm install @ruvector/attention

# Browser (WASM)
npm install @ruvector/math-wasm @ruvector/attention-wasm

Release v2.0.4

RuVector Release v2.0.4

Published Packages

crates.io

• ruvector-math - Advanced math primitives
• ruvector-attention - 7-theory attention mechanisms
• ruvector-math-wasm - WASM bindings for math
• ruvector-attention-wasm - WASM bindings for attention

npm

• @ruvector/math-wasm - Browser WASM package
• @ruvector/attention - Main Node.js package (auto-selects platform)
• @ruvector/attention-wasm - Browser WASM package
• Platform-specific: linux-x64, linux-arm64, darwin-x64, darwin-arm64, win32-x64

Installation

# Rust
cargo add ruvector-math ruvector-attention

# Node.js (auto-selects correct binary)
npm install @ruvector/attention

# Browser (WASM)
npm install @ruvector/math-wasm @ruvector/attention-wasm

Release v0.3.0

RuVector Release v0.3.0

Published Packages

crates.io

• ruvector-math - Advanced math primitives
• ruvector-attention - 7-theory attention mechanisms
• ruvector-math-wasm - WASM bindings for math
• ruvector-attention-wasm - WASM bindings for attention

npm

• @ruvector/math-wasm - Browser WASM package
• @ruvector/attention - Main Node.js package (auto-selects platform)
• @ruvector/attention-wasm - Browser WASM package
• Platform-specific: linux-x64, linux-arm64, darwin-x64, darwin-arm64, win32-x64

Installation

# Rust
cargo add ruvector-math ruvector-attention

# Node.js (auto-selects correct binary)
npm install @ruvector/attention

# Browser (WASM)
npm install @ruvector/math-wasm @ruvector/attention-wasm

RVF CLI rvf-v0.1.0

RVF CLI Release

Standalone vector database CLI for creating, querying, and managing RVF stores.

Download

Platform	Binary
Linux x64	rvf-linux-x64
Linux ARM64	rvf-linux-arm64
macOS x64 (Intel)	rvf-darwin-x64
macOS ARM64 (Apple Silicon)	rvf-darwin-arm64
Windows x64	rvf-windows-x64.exe

Quick start

# Download (macOS ARM64 example)
curl -L -o rvf https://github.com/ruvnet/ruvector/releases/download/rvf-v0.1.0/rvf-darwin-arm64
chmod +x rvf

# Create a store and query
./rvf create mydb.rvf --dimension 128 --metric cosine
./rvf status mydb.rvf

See the CLI README for full documentation.

Release v0.1.32

RuVector Release v0.1.32

Published Packages

crates.io

• ruvector-math - Advanced math primitives
• ruvector-attention - 7-theory attention mechanisms
• ruvector-math-wasm - WASM bindings for math
• ruvector-attention-wasm - WASM bindings for attention

npm

• @ruvector/math-wasm - Browser WASM package
• @ruvector/attention - Main Node.js package (auto-selects platform)
• @ruvector/attention-wasm - Browser WASM package
• Platform-specific: linux-x64, linux-arm64, darwin-x64, darwin-arm64, win32-x64

Installation

# Rust
cargo add ruvector-math ruvector-attention

# Node.js (auto-selects correct binary)
npm install @ruvector/attention

# Browser (WASM)
npm install @ruvector/math-wasm @ruvector/attention-wasm

Release v0.1.31

RuVector Release v0.1.31

Published Packages

crates.io

• ruvector-math - Advanced math primitives
• ruvector-attention - 7-theory attention mechanisms
• ruvector-math-wasm - WASM bindings for math
• ruvector-attention-wasm - WASM bindings for attention

npm

• ruvector-math-wasm - Browser WASM package
• @ruvector/attention - Main Node.js package (auto-selects platform)
• @ruvector/attention-wasm - Browser WASM package
• Platform-specific: linux-x64, linux-arm64, darwin-x64, darwin-arm64, win32-x64

Installation

# Rust
cargo add ruvector-math ruvector-attention

# Node.js (auto-selects correct binary)
npm install @ruvector/attention

# Browser (WASM)
npm install ruvector-math-wasm @ruvector/attention-wasm

v0.1.4: chore: Update NAPI-RS binaries for all platforms

  Built from commit b5b4858a26debd612e5ddbc80964d048dba401ff

  Platforms updated:
  - linux-x64-gnu
  - linux-arm64-gnu
  - darwin-x64
  - darwin-arm64
  - win32-x64-msvc

  🤖 Generated by GitHub Actions