devwithjev
— reading now— views
Submit a build

CUDA Inference Backend for Open-Jev

open-jev-fast provides a specialized CUDA inference backend for Open-Jev-27B with a drop-in Jev-compatible server, prefix-tree reuse, and checked benchmark artifacts.

lyuyiqi@lyuyiqi
Specialized CUDA inference backend for Open-Jev-27B with a drop-in Jev-compatible server, prefix-tree reuse, and checked benchmark artifacts.
Sep 30, 2026GitHubView on GitHub

Also filed under Tools & apps

  • Local-First Coding Task Model Selector

    Turnhelm is a local-first CLI that uses Jev or Laya to select the model and reasoning effort for each coding task.

  • Filters Context with Jev Relevance Questions

    The Oh-my-pi extension asks Jev batched Noul relevance questions about tool-output and earlier assistant-text chunks before they enter main-model context. It supports opt-in filtering and recoverable originals.

  • Jev Hub Tracker and Playground

    Jev Hub is a bilingual Jev tracker and interactive playground. Its templates submit typed Choice, Score, and Noul questions through a Cloudflare proxy and display structured results.

  • Jev Cross-Platform Chat Assistant

    Goutoujunshi Jev Chat is a cross-platform chat assistant. Jev chooses a bounded reply strategy from reviewed conversation context before the user sends any draft.