reposource-backed
Jev Search
Jev chooses sources and query settings, then re-ranks results returned by search engines. It does not generate an answer.
SELECTED DEMO PREVIEWS
See it
in use.
8 selected interfaces and recorded runs. Explore a preview, read the evidence, then open the author's demo.
These are attributed stills, captured screenshots and recorded replays. Each card keeps its evidence label. Media policy → · Selection
demosource-backed
Jev Trader: one decision per Monad block
A bot that asks Jev for a buy/sell decision on every Monad block from a price feed and executes on Kuru's MON/USDC book, with a live status page.
demosource-backed
Component Charades: Jev guesses a UI component
A Taboo-style game where a player describes a UI component without its forbidden words and Jev ranks the component library. A second mode finds a component from the job the player needs it to perform.
demosource-backed
Jev × LIBERO: robot decisions with local physics previews
A configurable LIBERO robot-control experiment that combines Jev decisions with reversible physics previews. An interactive replay follows intent, motion family and control choices alongside saved trajectories.
replicasource-backed
minojev: a trained decision head on a frozen language model
An independent Jev-style project that freezes a language-model backbone and trains a small decision head over cached candidate features. It publishes typed probability outputs, datasets and browser presentations of its benchmark and game policies.
benchmarksource-backed
YC × Jev: startup search and a reranking benchmark
An independent reranking benchmark paired with a live natural-language search over Y Combinator companies. The app uses Jev for query routing and reranking after keyword and embedding retrieval.
demoauthor demo
JevForm: a form that decides what to ask next
A dynamically branching form where Jev chooses the next question, built on Vercel's json-render for generative form UI and xstate for state — replacing hand-written if/then logic.
replicasource-backed
PlayJev: a 0.8B visual decision model for ten games
An independent Jev-like model fine-tuned from Qwen3.5-0.8B-Base to choose legal actions from raw game frames. The project publishes model weights, training records and a gallery of ten browser games.