[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-2936":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":16,"stars30d":16,"stars90d":16,"forks30d":16,"starsTrendScore":16,"compositeScore":17,"rankGlobal":10,"rankLanguage":10,"license":18,"archived":19,"fork":19,"defaultBranch":20,"hasWiki":21,"hasPages":19,"topics":22,"createdAt":10,"pushedAt":10,"updatedAt":43,"readmeContent":44,"aiSummary":45,"trendingCount":16,"starSnapshotCount":16,"syncStatus":46,"lastSyncTime":47,"discoverSource":48},2936,"ovo-local-llm","ovoment\u002Fovo-local-llm","ovoment","A private Claude-Code-style coding agent for Apple Silicon — run chat, code, and local model workflows on-device. MLX-native, Ollama\u002FOpenAI API compatible, zero API keys.","https:\u002F\u002Fovoment.com",null,"TypeScript",102,18,1,9,0,43.84,"MIT License",false,"main",true,[23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42],"agentic-ide","apple-silicon","claude-code","deepseek","diffusers","gemma","huggingface","llama","llm","lm-studio-alternative","local-llm","macos","mistral","mlx","ollama-alternative","on-device-ai","privacy-first","qwen","rust","tauri","2026-06-12 04:00:16","\u003Cp align=\"center\">\n  \u003Cimg src=\"docs\u002Fimages\u002Flogo.jpg\" alt=\"OVO — ovo-local-llm\" width=\"480\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fovoment\u002Fovo-local-llm\u002Fstargazers\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fovoment\u002Fovo-local-llm?style=flat&logo=github&color=yellow\" alt=\"Stars\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fovoment\u002Fovo-local-llm\u002Freleases\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fdownloads\u002Fovoment\u002Fovo-local-llm\u002Ftotal?logo=github&color=brightgreen\" alt=\"Downloads\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fovoment\u002Fovo-local-llm\u002Freleases\u002Flatest\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fv\u002Frelease\u002Fovoment\u002Fovo-local-llm?logo=github&color=blue\" alt=\"Release\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fovoment\u002Fovo-local-llm\u002Fissues\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues\u002Fovoment\u002Fovo-local-llm?logo=github\" alt=\"Issues\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fko-fi.com\u002Fovoment\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSupport%20on-Ko--fi-ff5e5b?logo=ko-fi&logoColor=white\" alt=\"Ko-fi\">\u003C\u002Fa>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-MIT-blue.svg\" alt=\"License\">\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FmacOS-13%2B-black?logo=apple&logoColor=white\" alt=\"macOS 13+\">\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FApple%20Silicon-M1%20%E2%86%92%20M4-orange\" alt=\"Apple Silicon\">\n\u003C\u002Fp>\n\n\u003Ch3 align=\"center\">🦉 A private Claude-Code-style coding agent for Apple Silicon\u003C\u002Fh3>\n\n\u003Cp align=\"center\">\n  Run chat, code, and local model workflows on-device — without shipping your work to the cloud.\u003Cbr>\n  MLX-native. Ollama\u002FOpenAI API compatible. Zero API keys.\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"README.ko.md\">🇰🇷 한국어 README\u003C\u002Fa>\n\u003C\u002Fp>\n\n---\n\n\u003Cp align=\"center\">\n  \u003Cvideo src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fff9f97cc-7a5c-4627-a547-d060ae1f2b86\" width=\"860\" controls autoplay muted loop>\n    Your browser does not support the video tag.\n  \u003C\u002Fvideo>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"docs\u002Fimages\u002Fchat.png\" alt=\"OVO chat\" width=\"860\">\n\u003C\u002Fp>\n\n## What's New in v0.0.6\n\n**Document parsing & RAG** — PPTX, HWP, HWPX parsing via kordoc. OCR for scanned PDFs. Knowledge Base auto-injects relevant docs into chat context.\n\n**LoRA fine-tuning** — Dataset creation, training, adapter merge. Full `mlx-lm` pipeline with configurable rank\u002Flayers\u002Fepochs.\n\n**Model blending** — Merge two same-architecture models via SLERP, Linear, TIES, or DARE. Progress tracking GUI included.\n\n**Ping Pong upgrades** — File\u002Fcode\u002FURL attachments, editor-style code blocks (dark bg + line numbers), auto-split long responses into multiple bubbles, typing indicator, thinking content stripped from output.\n\n**Other changes** — Web search only fires on explicit request. Settings page grouped into card sections. File extraction limit raised to 200 KB. Mac chip name shown in hardware cards.\n\n---\n\n## Code — a local coding agent with hands\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"docs\u002Fimages\u002Fcode.png\" alt=\"OVO Code IDE\" width=\"860\">\n\u003C\u002Fp>\n\nMonaco editor + file explorer + Git panel + PTY terminal + AI inline completion. The Agent Chat on the right gets file read\u002Fwrite\u002Fsearch\u002Fexec tools and MCP server integration — it can actually do the work, not just describe it. Like Claude Code, but running on your Mac with any open LLM.\n\n## Also included\n\n### Chat — every open LLM, one interface\n\nNative Ollama\u002FOpenAI API compatibility, streaming responses, persona switching, file attachments (PDF \u002F Excel \u002F Word \u002F images), voice input + TTS.\n\n### Image generation — diffusion on your laptop\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"docs\u002Fimages\u002Fimage.png\" alt=\"OVO image generation\" width=\"860\">\n\u003C\u002Fp>\n\nLocal text-to-image via `diffusers`. Sampler \u002F steps \u002F CFG \u002F LoRA controls.\n\n### Wiki — persistent knowledge across sessions\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"docs\u002Fimages\u002Fwiki.png\" alt=\"OVO wiki\" width=\"860\">\n\u003C\u002Fp>\n\nCurated notes + auto-captured session logs with BM25 + semantic search. Context that survives restarts.\n\n### Models — HuggingFace-native, zero re-downloads\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"docs\u002Fimages\u002Fmodels.png\" alt=\"OVO models tab\" width=\"860\">\n\u003C\u002Fp>\n\nAuto-detects `~\u002F.cache\u002Fhuggingface\u002Fhub\u002F` + LM Studio cache — models you already have just show up. Download from URL or search HuggingFace directly.\n\nOVO is model-agnostic. Recommended by origin:\n- **US** — Llama (Meta), Gemma (Google), Phi (Microsoft)\n- **EU** — Mistral \u002F Mixtral (Mistral AI, France)\n- **Asia** — Qwen (Alibaba), DeepSeek, GLM\n\nAll models run fully local — no data leaves your machine.\n\n### Hardware fit — pick a model that actually runs\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"docs\u002Fimages\u002Fhardwarefit.png\" alt=\"OVO hardware fit\" width=\"860\">\n\u003C\u002Fp>\n\nScores every model against your RAM \u002F GPU \u002F context headroom. Know before you download.\n\n### Desktop mascot\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"docs\u002Fimages\u002Fpet1.png\" alt=\"Owl — thinking\" width=\"320\">\n    \n  \u003Cimg src=\"docs\u002Fimages\u002Fpet2.png\" alt=\"Owl — idle\" width=\"320\">\n\u003C\u002Fp>\n\nAn SVG owl that reacts to your coding state. Double-click to summon the main window.\n\n## Install\n\n1. Download the latest `OVO_x.y.z_aarch64.dmg` from [**Releases**](https:\u002F\u002Fgithub.com\u002Fovoment\u002Fovo-local-llm\u002Freleases).\n2. Open the DMG and drag **OVO.app** onto the **Applications** shortcut.\n3. Back in the DMG window, double-click **`Install OVO.command`**.\n   It shows you exactly the one command it will run, you click **Run**, done.\n\nThat's it — no Terminal required.\n\n\u003Cdetails>\n\u003Csummary>Why the third step? (click to expand)\u003C\u002Fsummary>\n\nOVO's build is not yet signed with an Apple Developer ID (the $99\u002Fyr\nmembership is on the roadmap — see the milestone in [Issues](https:\u002F\u002Fgithub.com\u002Fovoment\u002Fovo-local-llm\u002Fissues)).\nWithout a signature, macOS flags the app with `com.apple.quarantine` and\nrefuses to launch it with the classic *\"OVO is damaged and can't be opened\"*\ndialog.\n\n`Install OVO.command` runs a single command to clear that flag:\n\n```bash\nxattr -rd com.apple.quarantine \u002FApplications\u002FOVO.app\n```\n\nNo `sudo`, no network, no background processes. The script is short and\nauditable — read it here before running:\n[scripts\u002Fdmg-templates\u002FInstall OVO.command](scripts\u002Fdmg-templates\u002FInstall%20OVO.command)\n\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>Prefer to do it by hand?\u003C\u002Fsummary>\n\n```bash\nxattr -rd com.apple.quarantine \u002FApplications\u002FOVO.app\nopen \u002FApplications\u002FOVO.app\n```\n\nIf your `\u002FApplications\u002FOVO.app` happens to be owned by `root` (rare on\nrecent macOS), prefix with `sudo`.\n\n\u003C\u002Fdetails>\n\n**First launch** bootstraps a Python runtime into `~\u002FLibrary\u002FApplication Support\u002Fcom.ovoment.ovo\u002Fruntime\u002F` (≈1.5 GB, ~3 min, one-time). Subsequent launches are instant.\n\n### System requirements\n\n- macOS **13+** on Apple Silicon (M1 \u002F M2 \u002F M3 \u002F M4). Intel Macs are not supported.\n- **16 GB RAM** minimum (7B Q4 models only); **32 GB+** recommended for 14B and above.\n- **10 GB** free disk for runtime + a couple of models.\n\n> **16 GB users:** Only 7B quantized (Q4) models run comfortably. Disable extra features in **Settings → Feature Flags** (Wiki, Skills, MCP) to keep the system prompt lean and maximize response speed. The **Hardware Fit** tab shows which models actually fit your machine.\n\n## Quick start\n\n1. Launch OVO.\n2. Go to **Models**, pick a model (Qwen3, Llama 3.3, Gemma, Mistral, DeepSeek, …), click download.\n3. Open **Chat** and send a message — the local model answers, no network calls.\n4. Open a project folder in **Code** to use the IDE + Agent Chat.\n\n## API compatibility\n\n| Flavor | Port | Use case |\n|--------|:----:|----------|\n| Ollama | `11435` | Drop-in replacement for Ollama clients (Open WebUI, Page Assist, …) |\n| OpenAI | `11436` | Point any OpenAI SDK at `http:\u002F\u002Flocalhost:11436\u002Fv1` |\n| Native | `11437` | OVO-specific endpoints — model management, Wiki, streaming, voice |\n\n## Claude Code integration (opt-in)\n\nOVO can **read** your local Claude Code config so the same context reaches your local model:\n\n- `CLAUDE.md` — injected as system context\n- `.claude\u002Fsettings.json` — preferences honoured\n- `.claude\u002Fplugins\u002F**` — behaviour hints\n\nDisabled by default. Flip it on in **Settings → Claude Integration**. OVO never touches claude.ai, API keys, session tokens, or anything that could affect your Claude account.\n\n## Development\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fovoment\u002Fovo-local-llm.git\ncd ovo-local-llm\n\n# frontend + Rust deps\nnpm install\n\n# Python sidecar venv (dev uses $HOME cache, avoids SMB locks)\ncd sidecar && uv sync && cd ..\n\n# run the full stack in dev mode\nnpm run tauri dev\n```\n\nRelease build: `npm run tauri build` — produces `.app` + `.dmg` under your Cargo target dir.\n\nDeeper docs: [docs\u002FARCHITECTURE.md](docs\u002FARCHITECTURE.md) · [docs\u002FARCHITECTURE.en.md](docs\u002FARCHITECTURE.en.md) · [docs\u002Frelease\u002FBUILD.md](docs\u002Frelease\u002FBUILD.md) · [docs\u002Frelease\u002FSECURITY.md](docs\u002Frelease\u002FSECURITY.md) · [docs\u002Frelease\u002FPRIVACY.md](docs\u002Frelease\u002FPRIVACY.md)\n\n## Architecture\n\n- **Shell** — Tauri 2 (Rust)\n- **Frontend** — React 18 + TypeScript + Tailwind + shadcn\u002Fui + Monaco\n- **Sidecar** — Python 3.12 FastAPI, spawned by Rust, user-cached venv bootstrapped with a bundled `uv`\n- **Runtimes** — `mlx-lm`, `mlx-vlm`, `mlx-whisper`, `transformers`, `diffusers`\n- **Storage** — SQLite (chats + Wiki), local filesystem (attachments, models)\n\n## Support\n\nOVO is a solo-developer project. Every coffee funds one more model architecture I can patch and support.\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fko-fi.com\u002Fovoment\">\u003Cimg src=\"https:\u002F\u002Fko-fi.com\u002Fimg\u002Fgithubbutton_sm.svg\" alt=\"Support me on Ko-fi\">\u003C\u002Fa>\n\u003C\u002Fp>\n\n## License\n\n[MIT](LICENSE) — use it, fork it, ship it.\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"docs\u002Fimages\u002Finfo.png\" alt=\"OVO about\" width=\"720\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  Made with 🦉 by \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fovoment\">ben @ ovoment\u003C\u002Fa>\n\u003C\u002Fp>\n","ovoment\u002Fovo-local-llm 是一个专为 Apple Silicon 设计的私有 Claude-Code 风格的编码代理，能够在设备上运行聊天、代码和本地模型工作流，无需将数据上传到云端。该项目采用 TypeScript 编写，支持 MLX-native 和 Ollama\u002FOpenAI API 兼容性，且无需任何 API 密钥。其核心功能包括文档解析与检索增强生成（RAG）、LoRA 微调、模型融合以及改进的交互体验等。适用于需要在本地进行开发工作并重视隐私保护的 macOS 用户，特别是那些使用 M1 至 M4 芯片的设备。",2,"2026-06-11 02:51:47","CREATED_QUERY"]