Playground
CommunityJev Demos & Experiments
Jev Calibration game
Sixteen true-or-false statements you answer, then Jev answers with a Noul probability. Keeping only its confident calls is the point of the game.
- Format
- Playground
- Reading time
- 2 min
- Published by
- Community
- Last verified
- Sep 2026
- Platforms
- Browser
Highlights
- You answer 16 true-or-false statements. The first shown is whether a tomato is botanically a fruit.
- Jev answers each one with a Noul. The sample reads P(true) and sets confidence to max(p, 1-p).
- The page says keeping only the high-confidence answers pushes accuracy toward 100%.
- A sample decision on the page: the Great Wall claim, answered no at 6% in 191 ms.
Watch out
Independent site, not TypeSafe, and not the AnthusAI calibration study. The accuracy claim is the site's description of the demo. The sample client posts to jevtypesafeai.com/api/v1/decide.
Questions about this entry
Short answers drawn from the linked project or page.
The game drops them. You sort Jev's 16 answers by confidence, keep only the top ones, and report accuracy on what survives. That is selective prediction by construction, so the rising number describes a self-selected subset, not performance on all 16. The page offers no full-set accuracy to compare against.
No. The sample just reads P(true) from the noul answer and sets confidence to max(p, 1-p), then leaves the cut-off to you. There is no threshold sweep, no reliability diagram and no statement of how many of the 16 you must keep. Sixteen general-knowledge statements are also too few to establish calibration either way.
That single example is a decision display, not evidence. The page shows one noul answering no at 6% in 191 ms at about $0.000009, but never plots predicted probability against observed frequency. It is also an independent site rather than TypeSafe, so the accuracy claim is the site's description of its own demo.
The page argues from human behaviour rather than a head-to-head: people are most certain exactly where common misconceptions live, like bulls and red, or the 10% of brain claim. That is a claim about human calibration, and no comparison against a regular LLM asking for its own confidence is published here.
Sixteen true-or-false statements; you answer each one first, then Jev answers the same statement with a Noul probability, and the first shown is whether a tomato is botanically a fruit.
The sample reads P(true) from the noul answer and sets confidence to max(p, 1-p), so a very low probability counts as high confidence that the statement is false rather than low confidence.
No. The entry records this as an independent site, not TypeSafe and not the AnthusAI calibration study, and the claim that keeping high-confidence answers pushes accuracy toward 100% is the site's own description of its demo.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
Hacker News: Software Bombe and live Enigma demo where Jev ranks cribs and judges plaintext.
ag_212 pointsShow HN: Jev as the Bletchley Park Analyst: A Software Bombe on EnigmaRead the thread on Hacker NewsHacker News: Show HN for public Minecraft PvP where a diffusion LM picks moves from structured state.
2 pointsShow HN: Minecraft PvP against DiffusionGemma-as-Jev, 24 ms per decisionRead the thread on Hacker NewsHacker News: Show HN: Jev finished a Pokémon Red run, with the Hall of Fame stats on the landing page.
pancomplex272 pointsShow HN: Jev Plays Pokémon RedRead the thread on Hacker NewsHacker News: Show HN: a mouse colony where each next move is Jev or a fixed rule.
1 pointShow HN: Jev-mice – mouse colony simulation using Jev and deterministic engineRead the thread on Hacker NewsMore Jev resources like this
A static page that races three LLM judging gates (Laya, JEV, and zero-token ZTC) on the same Tetris board with identical pieces, proposals, and noise, publishing paired-match results and the measured constants behind them.
Jev Demos & ExperimentsPlayground#community#playground#demo
Community
A 25-second IT-buzzword typing game built to feel how fast and cheap Jev is: each word gets five typed judgments in one request, and the screen shows the running cost next to the real HTTP request and response.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
pochang6/jev-buzzword-rushJavaScript
pochang6/jev-buzzword-rushJavaScriptAn interactive ZeroGPU demo of AutoTrust's JEV-9B: System 1 answers noul, choice, and score questions in one prefill pass with calibrated probabilities, while an Escalate tab hands low-confidence cases to the untouched Qwen3.5-9B base.
Jev Demos & ExperimentsPlayground#community#playground#demo
Community
AutoTrust's live game-arena demo: watch JEV-27B play Mario, a 3D Rubik's Cube, and Tetris while a side panel reports action probabilities and per-decision inference time, alongside a decision lab and a ViZDoom arena.
Jev Demos & ExperimentsPlayground#community#playground#demo
Community
A Gradio playground for KaLM-Jev Small structured judgments: choice, score, noul, and mixed JSON questions return probability tables and full responses without generating answer text on shared ZeroGPU.
Jev Demos & ExperimentsPlayground#community#playground#demo
Community
A Gradio demo of Maincode's 26B MATILDA-jev-v1 one-pass decision model: feed a text or JSON state, optionally an image, and a choice, noul, or score question to get calibrated probabilities over your options.
Jev Demos & ExperimentsPlayground#community#playground#demo
Community
From 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.