Materializr
開源參數式 3D CAD 給創客 — 受限草圖、實體建模、螺紋、SVG 與文字雕刻、STL/STEP 匯出。Android、iPad、Linux、Mac、Windows。Vibe 編碼。
AI 翻譯
尚未偵測到更新。GitHub 發布版本後會顯示在這裡。
個人 GitHub 帳號下公開儲存庫的擁有者可以登入並驗證所有權,自行管理介紹,無需管理員核准。
登入並產生驗證檔案,在 30 分鐘內將其提交至此儲存庫的預設分支。其他分支或分叉儲存庫中的檔案無效。然後在下方檢查檔案,即可無需管理員核准直接管理專案。請保留檔案以維持管理權限。
如需登入,將前往登入頁面。登入後請回到此處,再次點擊所選的驗證按鈕。
在專案資料夾中執行 /nomorevibe verify,即可驗證所有權並自行管理介紹。
一分鐘完成驗證製作者的介紹僅有上面的一句話。 更多詳情請閱讀儲存庫 README。 儲存庫 README ↗
正在載入完整 README。
檢視完整 README ↗A model and dataset hub built with Next.js 16. It browses everything on the Hugging Face Hub through the public API, hosts your users' own models and datasets, and runs chat and image models in a playground.
| Area | What it does | Built on |
|---|---|---|
| Models | Search, task filters, sort, pagination, model cards, file lists, "Use this model" code | Hugging Face Hub API |
| Datasets | Same as models, plus a preview table of the first rows of each split | Hub API + datasets-server |
| Community repos | Signed-in users publish models/datasets with a card and files | Supabase (Postgres + Storage) |
| Accounts | GitHub or email-link sign-in, profiles, saved repos | Supabase Auth |
| Playground | Streaming chat and text-to-image with any supported model | Hugging Face Inference Providers |
npm install
cp .env.example .env.local # fill in the keys below
npm run dev
Open http://localhost:3000. The hub works with no keys at all; accounts and the playground switch on as you add them.
Create a fine-grained token at https://huggingface.co/settings/tokens with Make calls to Inference Providers, and set HF_TOKEN. Runs are billed to that account, so the playground requires sign-in whenever accounts are on, and each user gets PLAYGROUND_RATE_PER_MIN runs per minute.
supabase/schema.sql. It creates profiles, repos, saves, the public repos storage bucket and all row-level security policies.https://YOUR_DOMAIN/auth/callback (and http://localhost:3000/auth/callback) to Redirect URLs.{{ .Token }}.NEXT_PUBLIC_HUB_SUPABASE_URL and NEXT_PUBLIC_HUB_SUPABASE_ANON_KEY.repos bucket: raise the file size limit if you expect large weights (the free plan caps uploads at 50 MB per file).Push to GitHub, import the repo in Vercel, add the same environment variables, deploy. Nothing else is required.
src/
app/
page.tsx home: search + trending
models/ datasets/ index pages and [...id] detail pages
c/[username]/[kind]/[slug]/ community repo page (+ owner settings)
u/[username]/ profile: repos and saves
new/ settings/ login/ publish, profile, sign-in
playground/ chat + image UI
api/chat api/image inference (streams through router.huggingface.co)
api/search header search: Hub quicksearch + community repos
lib/hf.ts Hub API client
lib/community.ts Supabase queries
proxy.ts keeps the auth session fresh (Next 16's middleware)
supabase/schema.sql
HF_ENDPOINT points the hub client at a mirror if you need one.revalidate: 300 in lib/hf.ts).