# VecturaKit

> Swift-based vector database for on-device RAG using MLTensor and MLX Embedders

- **URL**: https://www.freshcrate.ai/projects/VecturaKit
- **Author**: rryam
- **Category**: Databases
- **Latest version**: `6.1.0` (2026-04-30)
- **License**: MIT
- **Source**: https://github.com/rryam/VecturaKit
- **Language**: Swift
- **GitHub**: 283 stars, 30 forks
- **Registry**: github
- **Tags**: `mlx-swift`, `rag`, `swift`

## Description

Swift-based vector database for on-device RAG using MLTensor and MLX Embedders

## Recent releases

| Version | Date | Urgency | Changes |
| --- | --- | --- | --- |
| `6.1.0` | 2026-04-30 | High | ## What's Changed * Restore the NaturalLanguage-backed vectura-cli and TestExamples executable. * Add CLI README coverage and Codemagic smoke validation. * Add NLContextualEmbedder adapter coverage for vector queries. * Count file-backed documents from disk so partially loaded caches do not report stale totals.  **Full Changelog**: https://github.com/rryam/VecturaKit/compare/6.0.0...6.1.0 |
| `6.0.0` | 2026-04-29 | High | ## What's Changed * [codex] Remove swift-embeddings backend from core by @rudrankriyam in https://github.com/rryam/VecturaKit/pull/83   **Full Changelog**: https://github.com/rryam/VecturaKit/compare/5.3.0...6.0.0 |
| `5.3.0` | 2026-04-04 | High | ## What's Changed * Add VecturaOAIKit as a package target by @subsriram in https://github.com/rryam/VecturaKit/pull/81   **Full Changelog**: https://github.com/rryam/VecturaKit/compare/5.2.1...5.3.0 |
| `5.2.1` | 2026-04-02 | Medium | ## What's Changed * [codex] Pin swift-embeddings to stable 0.0.26 by @rudrankriyam in https://github.com/rryam/VecturaKit/pull/80   **Full Changelog**: https://github.com/rryam/VecturaKit/compare/5.2.0...5.2.1 |
| `5.2.0` | 2026-03-15 | Low | ## What's Changed * Add getDocument, documentExists API and fix delete no-op bug by @thatswiftguy in https://github.com/rryam/VecturaKit/pull/77  ## New Contributors * @thatswiftguy made their first contribution in https://github.com/rryam/VecturaKit/pull/77  **Full Changelog**: https://github.com/rryam/VecturaKit/compare/5.1.0...5.2.0 |
| `5.1.0` | 2026-03-05 | Low | ## What's Changed * Improve storage and BM25 search performance by @rudrankriyam in https://github.com/rryam/VecturaKit/pull/74 * Use bounded top-K selection to avoid full result sorting by @rudrankriyam in https://github.com/rryam/VecturaKit/pull/75 * Add realistic workload benchmark profile and optional CI stage by @rudrankriyam in https://github.com/rryam/VecturaKit/pull/76   **Full Changelog**: https://github.com/rryam/VecturaKit/compare/5.0.0...5.1.0 |
| `5.0.0` | 2026-02-12 | Low | ## Highlights - Add core SwiftEmbedder model-family resolution refactor and static embedding dimension handling improvements. - Add ModernBERT support in SwiftEmbedder. - Add RoBERTa and XLM-RoBERTa support in SwiftEmbedder. - Add embedder validation and BERT fallback robustness fixes. - Align package platform minimums with runtime requirements and clean up redundant availability guardrails.  ## Included pull requests - #69 Refactor SwiftEmbedder core model resolution - #70 Add ModernBE |
| `4.0.0` | 2026-02-05 | Low | ## Breaking Changes  - **VecturaMLXKit extracted to separate repository:** `VecturaMLXKit`, `VecturaMLXCLI`, and MLX-related tests have been moved to [rryam/VecturaMLXKit](https://github.com/rryam/VecturaMLXKit). This removes the need for Package Traits (SE-0450). - **Package Traits removed:** The `MLX` trait no longer exists. MLX users should add `VecturaMLXKit` as a separate dependency. - **swift-tools-version reverted to 6.0:** Swift 6.1 is no longer required for this package. - **mlx-swift-l |
| `3.0.0` | 2026-01-29 | Low | ## Highlights  - **Package Traits (SE-0450)**: MLX dependencies are now optional. Consumers using only `VecturaKit` or `VecturaNLKit` no longer download the heavy `mlx-swift-lm` package, significantly improving build times. - **Swift 6.1+ Required**: Bumped `swift-tools-version` to 6.1 for Package Traits support. - **Smaller Default Model**: Switched from `potion-retrieval-32M` to `potion-base-4M` for faster downloads and CI.  ## Breaking Changes  - **Swift 6.1+ / Xcode 16.3+ required** - **MLX |
| `2.5.3` | 2026-01-26 | Low | ## Highlights - Hardened add/index rollback to restore overwritten documents and keep storage/search consistent. - Stricter config/input validation (DB name, search options, embedder inputs); BM25 handles empty queries safely. - Performance suites now default to a deterministic embedder for faster runs (opt-in real embeddings via `VECTURA_PERF_USE_SWIFT_EMBEDDER=1`). - Updated MLX dependencies to latest mlx-swift-lm (2.30.3 / mlx-swift 0.30.3). - CLI vector dimension is auto-detected by default; |

## Dependency audit

- **Score**: 100/100
- **Total deps**: 0
- **Resolved**: 0
- **Unresolved**: 0
- **License conflicts**: 0
- **Warnings**: 0
- **Scanned**: 2026-05-11

## Citation

- HTML: https://www.freshcrate.ai/projects/VecturaKit
- Markdown: https://www.freshcrate.ai/projects/VecturaKit.md
- Dependencies JSON: https://www.freshcrate.ai/api/projects/VecturaKit/deps

_Generated by freshcrate.ai. Indexes github releases for AI-agent ecosystem packages._