{"id":50747782,"url":"https://github.com/deeflect/mcclaw","last_synced_at":"2026-06-10T22:30:52.889Z","repository":{"id":337199894,"uuid":"1152529329","full_name":"deeflect/mcclaw","owner":"deeflect","description":"Find which local LLMs actually run on your Mac. 340+ models, hardware-aware recommendations.","archived":false,"fork":false,"pushed_at":"2026-03-18T03:07:23.000Z","size":218,"stargazers_count":13,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-03-18T19:28:52.110Z","etag":null,"topics":["ai","apple-silicon","llm","local-ai","machine-learning","macos","ollama"],"latest_commit_sha":null,"homepage":"https://mcclaw.it.com","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/deeflect.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":null,"dco":null,"cla":null}},"created_at":"2026-02-08T02:23:47.000Z","updated_at":"2026-03-18T06:13:14.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/deeflect/mcclaw","commit_stats":null,"previous_names":["deeflect/mcclaw"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/deeflect/mcclaw","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deeflect%2Fmcclaw","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deeflect%2Fmcclaw/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deeflect%2Fmcclaw/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deeflect%2Fmcclaw/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/deeflect","download_url":"https://codeload.github.com/deeflect/mcclaw/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deeflect%2Fmcclaw/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34174148,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-10T02:00:07.152Z","response_time":89,"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","apple-silicon","llm","local-ai","machine-learning","macos","ollama"],"created_at":"2026-06-10T22:30:52.030Z","updated_at":"2026-06-10T22:30:52.884Z","avatar_url":"https://github.com/deeflect.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\n  \u003cimg src=\"assets/favicon.png\" width=\"80\" alt=\"McClaw logo\" /\u003e\n\u003c/p\u003e\n\n\u003ch1 align=\"center\"\u003eMcClaw\u003c/h1\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003cstrong\u003eFind which local LLMs actually run on your Mac\u003c/strong\u003e\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003ca href=\"https://mcclaw.it.com\"\u003e\u003cimg src=\"https://img.shields.io/badge/Live-mcclaw.it.com-black?style=flat-square\" alt=\"Live Site\" /\u003e\u003c/a\u003e\n  \u003cimg src=\"https://img.shields.io/badge/models-340%2B-blue?style=flat-square\" alt=\"340+ models\" /\u003e\n  \u003cimg src=\"https://img.shields.io/badge/platform-macOS-lightgrey?style=flat-square\" alt=\"macOS\" /\u003e\n  \u003cimg src=\"https://img.shields.io/badge/license-MIT-green?style=flat-square\" alt=\"MIT License\" /\u003e\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003ca href=\"https://mcclaw.it.com\"\u003eLive Site\u003c/a\u003e •\n  \u003ca href=\"https://blog.deeflect.com\"\u003eBlog\u003c/a\u003e •\n  \u003ca href=\"https://x.com/deeflectcom\"\u003eTwitter\u003c/a\u003e\n\u003c/p\u003e\n\n---\n\n![McClaw](assets/og-image.jpg)\n\n## About\n\nA tool that helps Mac users figure out which local LLMs can actually run on their hardware. You input your Mac Mini specs (chip and RAM) and get recommendations with performance estimates.\n\n### The Problem\n\nRunning LLMs locally on Mac is confusing:\n- \"Will this 70B model fit in my 32GB RAM?\"\n- \"What quantization should I use?\"\n- \"Which model is best for coding vs general chat?\"\n\nMost people download something too big, it crashes, and they give up.\n\n### The Solution\n\nA simple 3-step wizard:\n1. Select your Mac (M4/M4 Pro, RAM amount)\n2. Select your experience level\n3. See exactly which models fit your hardware\n\n![Setup wizard](screenshots/setup-wizard.jpg)\n\n## Tech Stack\n\n| Layer | Technology | Notes |\n|-------|------------|-------|\n| Frontend | React 18 + TypeScript | Stable, nothing fancy needed |\n| Animations | Framer Motion | Smooth wizard transitions |\n| Styling | Tailwind CSS 3 | Fast iteration |\n| Backend | Convex | Only for cloud model price comparisons |\n| Hosting | Vercel | Free tier handles everything |\n\n### Why No Server?\n\nThe model database is static data compiled into the frontend. This means:\n- Instant filtering with no API calls\n- Works offline after first load\n- Updates are just a redeploy\n\n## Model Database\n\n50+ models with multiple quantization variants. Some examples:\n\n| Model | Parameters | q4_k_m RAM | Category |\n|-------|------------|------------|----------|\n| Qwen 2.5 Coder 14B | 14B | 10.5 GB | Coding |\n| Llama 3.1 8B | 8B | 6.5 GB | General |\n| DeepSeek R1 32B | 32B | 22 GB | Reasoning |\n| LLaVA 7B | 7B | 6 GB | Vision |\n\nFull model compatibility table available in [data/models.md](data/models.md)\n\n### Device Configurations\n\n```typescript\ndevices = {\n  \"m4-16\":    { ram: 16, usableRam: 12 },  // ~4GB reserved for macOS\n  \"m4-24\":    { ram: 24, usableRam: 18 },\n  \"m4-32\":    { ram: 32, usableRam: 26 },\n  \"m4pro-48\": { ram: 48, usableRam: 40 },\n  \"m4pro-64\": { ram: 64, usableRam: 54 },\n}\n```\n\n## How Matching Works\n\n```typescript\n// Filter models that fit in available RAM\nmodels\n  .flatMap(model =\u003e model.variants)\n  .filter(variant =\u003e variant.ramGb \u003c= device.usableRam)\n  .sort(by benchmark scores, then by popularity)\n```\n\n![Results page](screenshots/results.jpg)\n\n## Design Notes\n\nThe UI is intentionally Apple-inspired:\n- Centered content with minimal chrome\n- Large, tappable buttons\n- Progress indicators\n- Clean typography\n\nThis feels familiar to Mac users and builds trust.\n\n### Progressive Disclosure\n\nBeginners see a curated \"Top Picks\" view with recommendations. Power users get the full table with all quantization options and benchmark scores.\n\n## Quantization Reference\n\n| Quantization | Quality | Size | Recommended Use |\n|--------------|---------|------|-----------------|\n| q4_k_m | ~95% | 50% | Default choice for most users |\n| q8_0 | ~99% | 100% | When quality matters |\n| fp16 | 100% | 200% | Benchmarking only |\n\n## Ideas for Later\n\n- MacBook support (currently Mac Mini only)\n- User-submitted performance reports\n- Direct Ollama integration to detect installed models\n\n---\n\nBuilt by [@deeflectcom](https://x.com/deeflectcom) for the [OpenClaw](https://openclaw.ai) community\n\nModel data sourced from Ollama, official papers, Open LLM Leaderboard, and personal testing on M4 Mac Mini 32GB.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdeeflect%2Fmcclaw","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdeeflect%2Fmcclaw","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdeeflect%2Fmcclaw/lists"}