What is the Model vs Market demo and what does it compare?

A site and demo repo where three AI decision models put a probability on any question after reading recent news; it compares those probabilities with live Polymarket and Kalshi prices and replays how each model changed its mind article by article.

How does Jev participate?

TypeSafe Jev is one of the three decision models, alongside OpenAI Decisions and Cloudflare Clef. Each model reads recent news, never the odds, and outputs a probability for every question; the digest does not show the exact Jev version or prompts.

How do I run it locally?

The README shows: pnpm install, then copy .env.example to .env.local and add your keys, then pnpm dev. A live deployment is at model-vs-market.vercel.app.

Which API keys does it need?

The excerpt lists OPENAI_API_KEY for the OpenAI Decisions API and VALYU_API_KEY for the search API, and says you add your keys in .env.local. The truncated table likely lists more keys for the other model providers.

Does it guarantee that models beat the market?

No. The README frames it as "the best way to test AI decision models: make them predict the future"; the digest contains no results showing either side winning, so no accuracy claim is made here.

What stack is Model vs Market built on?

A TypeScript web project run with pnpm, deployed on Vercel at model-vs-market.vercel.app, per the README and digest metadata (created 2026-10-07, pushed 2026-10-08, MIT license).

Who built Model vs Market?

GitHub user yorkeccak publishes the MIT-licensed repository; the digest shows 12 stars and 1 fork. The project's exact data-refresh cadence and market coverage are not stated in the excerpt.

Jev directory

CommunityJev Demos & Experiments

Model vs Market

A demo that has three decision models read recent news and price Polymarket and Kalshi questions, then compares their probabilities with live market odds and replays how each model changed its mind article by article.

See highlights Open github.com

Stars: 12
Language: TypeScript
License: MIT
Reading time: 2 min
Published by: Community
Author: yorkeccak
Last verified: Oct 2026

yorkeccak/model-vs-market

User repository on GitHub

View on GitHub

Three AI decision models read the news and price Polymarket and Kalshi questions against real money

Forks: 1
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

Tagscommunity github demo typescript prediction-markets

Works forJev for Scoring Jev for Evaluation Jev for Decision gating

Used inJev for Finance & Fintech

Languages: TypeScript
Platforms: Vercel AI Gateway Next.js
Capabilities: Open sourceRequires API key

Highlights

Three decision models (OpenAI Decisions, TypeSafe Jev, Cloudflare Clef) put probabilities on market questions.
Models read recent news and never see the odds, then the site compares them with live Polymarket and Kalshi prices.
Replays how each model changed its mind, article by article.
Live demo at model-vs-market.vercel.app; local run via pnpm install, copy the env file and pnpm dev.
Keys involved include OPENAI_API_KEY for the Decisions API and VALYU_API_KEY for the search API.

Quickstart

bash

pnpm install
cp .env.example .env.local   # add your keys
pnpm dev

Watch out

README and GitHub metadata reviewed 2026-10-17. The excerpt truncates the API-key table, so not every key is listed here, and there is no evidence of scoring accuracy, market coverage or update cadence beyond the live-site link.

Questions about this entry

Short answers drawn from the linked project or page.

From the Jev guides

Original write-ups on what this entry is for.

Per-Tick Decisions in Games

Sep 28, 2026

A game tick is a state plus a legal action list. How Clash Royale, Minecraft, and the earlier Mario, Pokémon, StarCraft, and drone loops ask Jev one step at a time.

Jev Demos & Experiments#gaming#agents#control

Read article

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai