{"id":49458638,"url":"https://github.com/keith-cy/melix","last_synced_at":"2026-04-30T08:06:18.712Z","repository":{"id":347131502,"uuid":"1192848732","full_name":"Keith-CY/melix","owner":"Keith-CY","description":"Local-first AI runtime for Apple Silicon with CLI and macOS operator workflows for LoRA training, benchmarking, and evaluation.","archived":false,"fork":false,"pushed_at":"2026-04-29T20:05:43.000Z","size":14455,"stargazers_count":3,"open_issues_count":13,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-04-29T20:28:32.269Z","etag":null,"topics":["ai-runtime","apple-silicon","inference-server","local-ai","local-llm","lora","lora-training","macos","mlx","mlx-lora","model-ops","qlora","swift"],"latest_commit_sha":null,"homepage":"","language":"Swift","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Keith-CY.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"docs/contributing.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2026-03-26T16:11:09.000Z","updated_at":"2026-04-29T20:05:50.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/Keith-CY/melix","commit_stats":null,"previous_names":["keith-cy/melix"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/Keith-CY/melix","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Keith-CY%2Fmelix","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Keith-CY%2Fmelix/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Keith-CY%2Fmelix/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Keith-CY%2Fmelix/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Keith-CY","download_url":"https://codeload.github.com/Keith-CY/melix/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Keith-CY%2Fmelix/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32458297,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-29T22:27:22.272Z","status":"online","status_checked_at":"2026-04-30T02:00:05.929Z","response_time":57,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-runtime","apple-silicon","inference-server","local-ai","local-llm","lora","lora-training","macos","mlx","mlx-lora","model-ops","qlora","swift"],"created_at":"2026-04-30T08:05:40.548Z","updated_at":"2026-04-30T08:06:18.704Z","avatar_url":"https://github.com/Keith-CY.png","language":"Swift","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\n  \u003cimg src=\"apps/macos-menubar/Sources/AppMain/Resources/Branding/melix_logo.svg\" alt=\"Melix logo\" width=\"180\"\u003e\n\u003c/p\u003e\n\n# Melix\n\n[![release gates](https://img.shields.io/github/actions/workflow/status/Keith-CY/melix/release-gates.yml?branch=main\u0026label=release%20gates)](https://github.com/Keith-CY/melix/actions/workflows/release-gates.yml)\n[![app packaging](https://img.shields.io/github/actions/workflow/status/Keith-CY/melix/package-self-contained-app.yml?branch=main\u0026label=app%20packaging)](https://github.com/Keith-CY/melix/actions/workflows/package-self-contained-app.yml)\n![platform](https://img.shields.io/badge/platform-Apple%20Silicon-black)\n![macOS](https://img.shields.io/badge/macOS-15%2B-000000)\n![Swift](https://img.shields.io/badge/Swift-6-F05138)\n![Python](https://img.shields.io/badge/Python-3.12%2B-3776AB)\n![focus](https://img.shields.io/badge/focus-LoRA%20%2B%20Bench%20%2F%20Eval-0F766E)\n[![license](https://img.shields.io/badge/license-Apache%202.0-blue)](LICENSE)\n\nMelix is a local-first AI runtime for Apple Silicon. It combines a Swift control plane, a Python worker stack, a public CLI, and a native macOS operator surface so one machine can manage local model serving, LoRA training, and benchmark or evaluation workflows from the same runtime.\n\n## What Melix Is\n\nMelix is not just a thin local inference wrapper. The current repository is aimed at a practical local model-operations loop:\n\n- import or download models into a local registry\n- bind those models to managed server sessions\n- run local chat and operator workflows through a shared control plane\n- train and activate LoRA or QLoRA adapters\n- benchmark, evaluate, compare, and export results from the same product surface\n\n## Why Melix Exists\n\nLocal AI work on Apple Silicon often gets split across too many disconnected tools: one script for serving, another for LoRA training, a notebook for evaluation, and an ad hoc shell history for benchmarks. Melix is being built to keep those loops together.\n\nThe current product direction especially favors LoRA training and benchmark discipline:\n\n- train adapters without leaving the local Melix workflow\n- compare base and derived models through the same CLI and operator UI\n- keep benchmark, matrix benchmark, and evaluation evidence in one repository-owned format\n- turn repeatable local benchmarking into part of the product, not an afterthought\n\n## Who It Is For\n\n- Apple Silicon practitioners who want a local runtime instead of a remote-only workflow\n- model engineers who need a repeatable loop for LoRA fine-tuning, comparison, and evaluation\n- local AI product builders who want a same-host runtime with both CLI and operator surfaces\n- contributors who care about typed protocols, reproducible runbooks, and productized local tooling\n\n## What You Can Do Today\n\n- manage model roots, inspect registry state, and download or import local models\n- create, select, start, pause, resume, wake, and stop server sessions\n- run local chat flows through the `melix` CLI and the macOS operator surface\n- train, activate, publish, and remove derived LoRA-backed models\n- run `bench`, `bench matrix`, `eval`, and `eval compare` workflows and export artifacts\n- package Melix for local launch agents, Homebrew service use, or preview app-bundle delivery\n\n## Quick Start\n\nFor the shortest repository setup path:\n\n```bash\nmake bootstrap\nmake proto\nmake swift-test\nmake py-test\nmake integration-test\n```\n\nThen use:\n\n- [Getting Started](docs/getting-started.md) for the local stack, operator app, and first CLI flows\n- [Current Status](docs/current-status.md) for what is actually shipped today\n- [Documentation Map](docs/README.md) for the deeper protocol, runbook, and plan structure\n\n## Documentation\n\n- [Getting Started](docs/getting-started.md)\n- [Current Status](docs/current-status.md)\n- [Contributing](docs/contributing.md)\n- [Docs Map](docs/README.md)\n- [Benchmark, Evaluation, and LoRA Runbook](docs/runbooks/benchmark-matrix-evaluation-and-lora.md)\n- [Local Install Runbook](docs/runbooks/phase-8-local-install.md)\n- [Packaging Targets](docs/runbooks/platform-packaging-targets.md)\n\n## Contributing\n\nContributions are welcome. Start with [docs/contributing.md](docs/contributing.md) for the repository workflow, expected verification commands, documentation rules, and handoff expectations.\n\nIf a change alters behavior, update the relevant spec, runbook, roadmap, or plan in the same change. The README should stay focused on the project itself; operational detail belongs under `docs/`.\n\n## License\n\nMelix is licensed under the Apache License, Version 2.0. See [LICENSE](LICENSE).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkeith-cy%2Fmelix","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkeith-cy%2Fmelix","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkeith-cy%2Fmelix/lists"}