User repository on GitHub
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
Jev directory
CommunityJev Tools & CLIs
Good Papers
A Rotten Tomatoes-style ratings site for ML papers: reader votes plus a panel of 20 AI reviewer personas that scores every paper from day one, with Jev answering 20 typed questions per paper and classifying areas and paper types.
- Stars
- 7
- Language
- TypeScript
- Reading time
- 2 min
- Published by
- Community
- Author
- LzyFischer
- Last verified
- Oct 2026
- Languages
- TypeScriptPython
- Platforms
- Next.jsVercel AI GatewayPostgreSQL
- Capabilities
- Open sourceSelf-hostedRequires API key
Highlights
- Readers upvote or downvote papers they have read; votes are weighted by reputation, with conflicts of interest left out.
- A panel of 20 reviewer personas — rigor, novelty, clarity, reproducibility, practical use — scores each paper before any votes arrive.
- Jev (TypeSafe AI) answers 20 typed questions per paper, one per persona, plus the research area among ~190 areas and the paper type.
- Requires Node.js 20+ and a Jev API key (TYPESAFE_API_KEY, JEV_MODEL); a Python worker labels comment stances with Jev too.
- All 6,000+ accepted NeurIPS 2026 papers are covered session by session; the site is live at goodpapers.org.
Quickstart
bash
git clone https://github.com/LzyFischer/good-papers.git
cd good-papers
npm install
cp .env.example .env.local
npm run devWatch out
README reviewed 2026-10-17; the digest excerpt truncates inside the screenshot section, so the Jev panel and architecture details here come from the same README that the digest draws from. The repository lists no licence, and the README gives no accuracy or cost figures for the AI panel.
Questions about this entry
Short answers drawn from the linked project or page.
A community ratings site for ML papers, live at goodpapers.org: readers vote on papers they have read, and a panel of 20 AI reviewer personas scores every paper from day one. The README describes it as Rotten Tomatoes for research papers.
Jev (TypeSafe AI) answers 20 typed questions per paper — one per reviewer persona — plus the research area among about 190 areas in 20 groups and the paper type. A Python discussion worker also labels comment stances with Jev.
Personas for rigor, novelty, clarity, reproducibility, practical use and similar qualities give each paper a score before any reader votes. AI reviewers also open a debate under every paper and every AI comment is labeled.
Node.js 20+ and Python 3.11+ for the discussion worker, a Supabase project, and a Jev API key set as TYPESAFE_API_KEY with JEV_MODEL. The README covers cloning, npm install, and applying supabase/schema.sql to a fresh project.
Popularity never stands in for quality, per the README: trending shelves follow Hugging Face's lists, but Must read is the year's highest-rated work, votes are weighted by reputation, and conflicts of interest are left out.
All 6,000+ accepted NeurIPS 2026 papers, session by session, plus papers discovered through OpenAlex, the arXiv API, and Hugging Face's trending lists. New papers are judged on first view or by a daily cron.
The repository is public and the site live, but the digest shows no licence field and no pricing. Running it yourself costs a Jev API key plus Supabase and Vercel hosting; the worker is optional and runs on GitHub Actions.
More Jev resources like this
A Chinese-language Next.js site that takes your situation and picks 3–5 fitting items from 671 suggestions in a life guide, ranked by Jev choice, noul and score questions with evidence grades, costs and sources shown.
Jev Tools & CLIs#community#github#nextjs
Community
zjl1985/life-decisionTypeScript
zjl1985/life-decisionTypeScriptAn open-source alternative to Grok Bot built with Mastra and Nebius Token Factory: your own team of AI bots in one chat app, each with its own model, tools, memory and cloud browser; Jev keys are optional and power room routing.
Jev Tools & CLIs#community#github#typescript
Community
Studio1-OSS/OpenCrewBotTypeScript
Studio1-OSS/OpenCrewBotTypeScriptA self-hostable HTTP proxy that fronts TypeSafe Jev through one endpoint, rotating your API keys automatically and serving the /v1/systemone wire format. Keys live in Upstash Redis or an environment variable, and a web admin page manages them.
Jev Tools & CLIs#community#github#typescript
Community
left0ver/jev-proxy-poolTypeScript
left0ver/jev-proxy-poolTypeScriptAn open-source, vendor-neutral web UI sandbox for /v1/systemone structured decision models such as TypeSafe Jev and Ollama clef: visualize Choice, Score and Noul probabilities, compare endpoints A/B, and export SDK calls, with a built-in mock that needs no network.
Jev Tools & CLIs#community#github#nextjs
Community
doggy8088/s1-studioTypeScript
doggy8088/s1-studioTypeScriptOpen-source AI email assistant (Next.js, Postgres, 12k+ stars) with a decision-model layer that can run on openrouter:typesafe/jev-1.13: Jev-backed questions cover rule selection, cold-email detection, thread status, recurring patterns, unsubscribe pages, and sender categorization, each with evals.
Jev Integrations & Adapters#community#email#gmail
Community
elie222/inbox-zeroTypeScript
elie222/inbox-zeroTypeScriptA search API over about 203 million OpenAlex English title and abstract records, running DenseOn embedding search by meaning and a tantivy index for exact words behind one HTTP service. Its whole-field topic maps use TypeSafe's Jev to judge archetype and cited papers.
Jev Tools & CLIs#community#github#python
Community
maximerivest/srl-searchPython
maximerivest/srl-searchPythonFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.