Hugging Face's logo

maxffarrell
/
gpu-search

English
gpu-search
experimental
embeddings
int8
typescript
navigation
Model card Files
xet
 Community
gpu-search / docs /deployment.md
maxffarrell's picture
maxffarrell
Upload current experimental gpu-search model with runtime and provenance
546d6ba verified
preview code
|
Raw History Blame Contribute Delete
1.13 kB

Deployment

Public demo: https://gpu-search.vercel.app

GitHub: https://github.com/maxffarrell/gpu-search

Published 2026-09-12 to Vercel project gpu-search in maxffarrells-projects. The Vercel team API reported billing plan hobby before project creation. No paid features or plan changes were enabled. The demo is static; it requires no secrets, server functions, database, analytics, or inference service.

The initial CLI production deployment dpl_BYWZ9EjyZYaZL8aZKLegvPk3G8K5 reached READY and was aliased to the public URL. Vercel connected the GitHub repository for subsequent builds. vercel.json sets the frozen pnpm install, Vite build, output directory and restrictive security headers. .vercelignore excludes research data and float checkpoints. The demo revision includes the experimental manifest and 32KiB int8 weights as explicit assets for real local CPU inference.

Run vercel deploy --prod --scope maxffarrells-projects to republish when explicitly intended. Runtime behavior was validated on the local production build; deployment readiness and the alias were verified from Vercel's build result.