Jev directory
CommunityJev Demos & Experiments
Turn any LLM into a Jev-like model
A live demo of a trained KV cache bank that, per its madewithjev listing, turns any LLM into a Jev-like model. The demo URL returned a URLError at fetch time, so the listing title and description are the only sources for this entry.
- Reading time
- 2 min
- Published by
- Community
- Author
- faangguyindia
- Last verified
- Oct 2026
Works forJev for Classification
Used inJev for Software & DevTools
Highlights
- Listed on madewithjev as a trained KV cache bank with a live demo that turns any LLM into a Jev-like model.
- The demo URL is a Tailscale ts.net host and returned a URLError on 2026-10-15, so no page text or commands were recoverable.
- The listing credits faangguyindia and gives no repository, license, hardware requirement or cost.
Watch out
Everything here comes from the madewithjev listing title and description fetched on 2026-10-15; the demo page returned a URLError, and no repo, model card or author page was linked, so the mechanism, requirements, license and status could not be verified.
Questions about this entry
Short answers drawn from the linked project or page.
A madewithjev listing for a live demo of a trained KV cache bank that turns any LLM into a Jev-like model. The listing credits faangguyindia and provides only a demo link, with no repository or write-up attached.
The listing does not explain it beyond the phrase 'a trained KV cache bank'. A key-value cache normally stores attention states for reuse; what is trained here, and how it makes an LLM behave like Jev, is not described in the available material.
The URL, a Tailscale ts.net host named gambler-relay-us-west1, returned a URLError when fetched for this batch on 2026-10-15. Tailscale hosts are typically reachable only on the owner's tailnet, so the demo may not be public.
Only through the listing's claim that the result behaves like Jev, TypeSafe's System One model that answers typed questions with probabilities instead of generating text. No compatibility with Jev's API or output shape is asserted or shown.
Unknown. No repository, model card or weights are linked in the listing, and no license is mentioned anywhere in the digest, so there is nothing to inspect or reuse from what was fetched.
The listing credits the handle faangguyindia. There is no further identity, date, changelog or status information in the digest, and the unreachable host means nothing could be checked live.
The catalog's other open reproductions are documented and reachable, such as openjev, decider and Jeff, while this item is only a pointer. If the tailnet host opens for you, treat it as a demo and verify the method before relying on it.
More Jev resources like this
A magic eight ball answered by Jev: ask a question and the decision model picks an answer in the shape of the classic toy. Built by sarreph and listed on madewithjev, the community directory of Jev builds.
Jev Demos & ExperimentsPlayground#community#catalog-import#madewithjev
Community
A local app that turns a worded query into a physics pile of floating YC logos: retrieval over 6,241 companies by meaning, logo image, OCR'd logo text, and Jev re-applied tags, with Jev scoring the finalists into calibrated probabilities.
Jev Demos & Experiments#community#catalog-import#madewithjev
Community
aayan-dev/aayans-yc-indexorHTML
aayan-dev/aayans-yc-indexorHTMLA Next.js console whose whole design system is generated at runtime by Jev from a one-sentence brief: type 'an 18th century printed book' and every visual decision is chosen, not written, from a catalog a person authored.
Jev Demos & Experiments#community#catalog-import#madewithjev
Community
bilune/jev-designTypeScript
bilune/jev-designTypeScriptAn experiment that drives Jev as an autoregressive next-token predictor: each step offers potential next tokens as choice options, and the resulting sequence of picks is the generated text. The author's verdict: it kinda works.
Jev Demos & Experiments#community#catalog-import#madewithjev
Community
erik-dunteman/chatjevPython
erik-dunteman/chatjevPythonFire-evacuation simulation where every occupant gets a Jev-style typed decision engine: a local Ternary Bonsai 2 27B, Qwen3.5-4B or the TypeSafe Jev API reads choice probabilities, while movement runs on validated social-force physics.
Jev Demos & Experiments#community#catalog-import#madewithjev
Community
using76/typeevacsafePython
using76/typeevacsafePythonA no-dependency single-page classifier that unpacks a Jev call: paste an API key, edit a choice question, and see the request JSON, per-category probabilities, confidence, latency, token usage and the equivalent curl.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
q93304989-bit/jev-labHTML
q93304989-bit/jev-labHTMLFrom 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.