{"id":48228697,"url":"https://github.com/lemberalla/manwe-releases","last_synced_at":"2026-07-02T14:00:48.987Z","repository":{"id":346889571,"uuid":"1190545221","full_name":"lemberalla/manwe-releases","owner":"lemberalla","description":"Multi-agent swarm intelligence for macOS","archived":false,"fork":false,"pushed_at":"2026-05-03T19:29:02.000Z","size":5798,"stargazers_count":5,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-03T21:27:32.361Z","etag":null,"topics":["ai","apple-silicon","decision-making","local-ai","macapp","macos","multi-agent","privacy","swiftui"],"latest_commit_sha":null,"homepage":"https://tinythings.app/manwe","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/lemberalla.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":"2026-03-24T11:43:17.000Z","updated_at":"2026-05-03T19:29:06.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/lemberalla/manwe-releases","commit_stats":null,"previous_names":["lemberalla/manwe-releases"],"tags_count":15,"template":false,"template_full_name":null,"purl":"pkg:github/lemberalla/manwe-releases","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lemberalla%2Fmanwe-releases","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lemberalla%2Fmanwe-releases/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lemberalla%2Fmanwe-releases/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lemberalla%2Fmanwe-releases/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/lemberalla","download_url":"https://codeload.github.com/lemberalla/manwe-releases/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lemberalla%2Fmanwe-releases/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35050017,"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":"2026-07-02T02:00:06.368Z","response_time":173,"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","decision-making","local-ai","macapp","macos","multi-agent","privacy","swiftui"],"created_at":"2026-04-04T19:28:31.677Z","updated_at":"2026-07-02T14:00:48.972Z","avatar_url":"https://github.com/lemberalla.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\n  \u003cimg src=\"assets/manwe-icon-beta.webp\" width=\"128\" height=\"128\" alt=\"Manwe\"\u003e\n\u003c/p\u003e\n\n\u003ch1 align=\"center\"\u003eAI doesn't need better answers. It needs better disagreement.\u003c/h1\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003cstrong\u003eManwe\u003c/strong\u003e is a decision room for macOS.\u003cbr\u003e\n  Describe a decision. Get a debate. Receive an auditable decision record.\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003ca href=\"https://github.com/lemberalla/manwe-releases/releases/latest\"\u003e\u003cstrong\u003eDownload the Beta\u003c/strong\u003e\u003c/a\u003e · \u003ca href=\"https://discord.gg/Nz6RExEpSD\"\u003eJoin the Discord\u003c/a\u003e · \u003ca href=\"https://askmanwe.com\"\u003eWebsite\u003c/a\u003e\n\u003c/p\u003e\n\n---\n\nDescribe a decision or question. Manwe searches 30+ source and evidence surfaces, assembles a panel of AI advisors with diverse perspectives, and runs a multi-round debate. You get a structured decision record with predictions, confidence scores, evidence, risks, and a concrete action plan. Not a chat response.\n\nYour advisors are alive. They remember past debates, develop worldviews over time, and transform when genuinely challenged. Between runs, they react to real-world knowledge in their domain — leave the app open and they grow.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"assets/screenshot-1.png\" width=\"100%\" alt=\"The Panel — advisors debating with unique perspectives\"\u003e\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"assets/screenshot-2.png\" width=\"100%\" alt=\"Structured decision record with predictions and confidence scores\"\u003e\n\u003c/p\u003e\n\n## Try asking\n\n- \"Should I leave my job to start a company? I'm 32, earning $140K, with 3 paying customers.\"\n- \"Will Bitcoin hit $200K by end of 2026?\"\n- \"My doctor recommended a Whipple procedure — what are the risks and alternatives?\"\n- \"I want to wash my car. The car wash is 50 meters away. Should I walk or drive?\"\n\n## Runs on your Mac\n\nRuns locally on Apple Silicon. Or connect your cloud provider. Your choice.\n\nManwe runs Qwen locally via MLX — no API keys, no data leaving your device. For a quality leap, connect Claude, Codex, or any OpenAI-compatible API. Bring your own key, bring your own model.\n\n| Model | Tier | Type | Requirements |\n|-------|------|------|-------------|\n| Qwen3 8B | Standard | Local | 8GB+ RAM, ~5GB download |\n| Qwen3.5 9B | Pro | Local | 16GB+ RAM, ~5.5GB download |\n| Qwen3.6 35B MoE | Ultra | Local | 36GB+ RAM, ~19GB download |\n| Qwen3.5 35B MoE | Ultra | Local | 36GB+ RAM, ~19GB download |\n| Claude Sonnet / Opus | Cloud | CLI | Claude Code CLI + subscription |\n| Codex (GPT) | Cloud | CLI | Codex CLI + ChatGPT/OpenAI subscription |\n| Qwen 3.6 Cloud | Cloud | CLI | Qwen Code CLI + authentication |\n| Anthropic / OpenAI / Alibaba API | Cloud | API Key | Your own API key |\n| OpenRouter, Groq, Together, Ollama... | Cloud | Custom | Any OpenAI-compatible endpoint |\n\n## What makes it different\n\n- **Chat first, run when ready** — talk a decision through, invite up to 3 advisors from your pool, then hit Run when you want a structured debate\n- **Market Intelligence** — market-aware runs connect fresh signals, filings, earnings context, price/volume history, and source-backed catalysts\n- **Event Radar** — scan markets, companies, technology, policy, health, climate, geopolitics, and culture before deciding whether to run a full debate\n- **Computed evidence cards** — timing context, earnings checks, market screens, and deeper evidence summaries open as inspectable cards\n- **Reasoning control** — pick how hard each model thinks. Low to Max, per model, for Claude, Codex, and API backends\n- **Your data, their debate** — upload documents (PDF, TXT, MD) and every advisor sees your facts. The Auditor fact-checks claims against your actual data, not just web search\n- **Real research, not vibes** — searches scholarly, legal, government, market, software, health, macro, and public-record sources before the debate starts\n- **Advisors who actually disagree** — the panel always includes a Contrarian (challenges consensus) and an Auditor (fact-checks claims)\n- **Living agents** — advisors develop worldviews, accumulate experience across runs, and react to real-world knowledge between debates\n- **Personalized evidence** — each advisor reads research ranked by relevance to their specific expertise\n- **Guest experts** — Manwe detects knowledge gaps and recruits specialists on the fly\n- **Claim verification** — when agents cite statistics, the system auto-searches and regenerates with real evidence\n- **You're in the room** — inject events mid-debate, interview individual advisors after\n- **Continue chains** — follow up on any completed decision record with new questions\n- **Foundation Models** — Apple's Neural Engine handles research tasks while the GPU runs debates\n- **Ask Manwe Web** — sign in at app.askmanwe.com for Quick and Pro web runs, saved records, source inspection, and usage controls\n\n## Install\n\n1. Download the latest DMG from [**Releases**](https://github.com/lemberalla/manwe-releases/releases/latest)\n2. Open the DMG and drag Manwe to Applications\n3. Launch — onboarding guides you through model setup\n\n**Requires macOS 14.0+ and Apple Silicon (M1 or later).**\n\n## Changelog\n\nSee [**Releases**](https://github.com/lemberalla/manwe-releases/releases) for the full version history.\n\n## Feedback\n\nThis is a beta. Things will break.\n\n**[Join the Discord](https://discord.gg/Nz6RExEpSD)** — bug reports, feature ideas, or just share your wildest decision records.\n\n---\n\n\u003cp align=\"center\"\u003e\n  Made by Oncel Cebeci · \u003ca href=\"https://tinythings.app\"\u003ea tiny things app\u003c/a\u003e\n\u003c/p\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flemberalla%2Fmanwe-releases","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Flemberalla%2Fmanwe-releases","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flemberalla%2Fmanwe-releases/lists"}