alexrozanski/LlamaChat
49.5
Weak · 1 October 2026
6.8k
lines of production code
Swift
primary language
2
measurements over time
What this system is
LlamaChat is a macOS application designed for interacting with local large language models such as LLaMA, Alpaca, and GPT4All. It provides a chat interface with message history persistence via SQLite, allowing users to manage multiple model sources and configure inference parameters like context size and temperature. The system includes utilities for version management and DMG packaging to streamline the release process.
Features
Added scripts for version bumping and DMG packaging
New shell scripts have been added to the Scripts directory to streamline the release process. The bump-version.sh script allows developers to increment the major, minor, or patch version numbers in the LlamaChat.xcconfig file, automatically updating both the marketing and project build versions. The make-dmg.sh script automates the creation of the application's disk image by verifying notarization status, generating the AppIcon, and using create-dmg to package the app with a custom background and layout. A .gitignore file has also been added to exclude generated .dmg files and temporary build artifacts from version control.
Scripts · high confidence
Initial project structure for Camel macOS app
The Camel application project has been initialized with a standard Xcode project file, establishing the core file structure for the app. This includes the main application entry point (CamelApp.swift), model classes for chat and messages, view models for managing chat, compose, and message states, and UI components for the main content view, chat list, message display, and composition interface. The project is configured to build a macOS application target named 'Camel' with associated assets and entitlements.
Camel.xcodeproj · high confidence
LlamaChat initial release with chat, settings, and model management
LlamaChat is now available as a standalone macOS application for chatting with local LLaMA, Alpaca, and GPT4All models. The app features a chat interface with message history persistence via SQLite, allowing users to manage multiple chat sources. Users can configure model parameters such as context size, temperature, and thread count, and the app supports custom avatars for different sources. Additional features include automatic cleanup of unquantized model files, text selection in message bubbles, and an integrated update mechanism via Sparkle.
LlamaChat · high confidence
Removals
Removed initial chat application scaffolding
The initial SwiftUI-based chat application structure has been removed, including the main app entry point, the chat model for managing message state, and the user interface components such as the content view, chat list, and message display views. This change eliminates the basic setup and initial chat flow that was previously present in the codebase.
camel · high confidence
Behavioural changes
App renamed to LlamaChat with updated project structure
The application has been renamed from CamelChat to LlamaChat, reflected in the project file, workspace, and entitlements. This change includes the addition of numerous source files for new features such as source conversion, model parameter settings, and Sparkle integration, alongside the replacement of the MessagesView with an NSTableView-backed implementation for selectable message text.
LlamaChat.xcodeproj · high confidence
Dependencies
Initial dependency resolution for LlamaChat
The project now includes a resolved package manifest defining its external dependencies. This establishes the specific versions for Coquille (v0.3), llama.swift (v2), Sparkle (2.4.1), and SQLite.swift (0.14.1), ensuring consistent builds across the development environment.
(dependencies) · high confidence
Written by watchdog.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.
How this codebase got here
Score
- CAI 55 → 49 (-5.9)
- Rubric changed (rubric-2026.09.11 → rubric-2026.09.18) — scores are not directly comparable.
Lenses
- Code Health 99 → 98 (-0.5)
- Architecture 100 → 100 (-0.4)
- Maturity 51 → 51 (+0.0)
- Readiness 31 → 21 (-10.1)
- Security 97 → 97 (+0.0)
Resolved (3)
- Dependency hygiene PARTLY measured — SwiftPM pinning read, dependency currency NOT established
- Documentation: no contributor guidance (README.md)
- Documentation: no licence statement (README.md)
New (5)
- Duplicated block (15 lines × 2) (LlamaChat/ui/components/AvatarPickerView.swift)
- Duplicated block (9 lines × 2) (LlamaChat/ui/settings/tabs/sources/detail/SourceSettingsParametersView.swift)
- Outdated: sparkle
- Outdated: sqlite.swift
- StateRestoration.swift.toRestorableValue (cognitive 17) (LlamaChat/ui/restoration/StateRestoration.swift)
Written by watchdog.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.
Survey your own repository
alexrozanski/LlamaChat was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point a surveyor at a repository you know and see whether you agree with it.
About this page
- The score is its most recent published measurement, taken on 1 October 2026 at a pinned commit. It is not a live figure and does not change until the project is measured again.
- Measured at commit 2b1f04bc07ba17a1877ca96ed9f4d767b769e63e — the exact code this score is about.
- Scored under rubric-2026.09.18 — the same rubric and the same method as every other entry in this index.
- Measured by watchdog.canine.dev using codehealth-analyzer preprod-e569280dd5e2.