CCommunity contributorResources

Verdict (Open-jev) launch thread

Hemant Kumar on his 151M ModernBERT reproduction: 0.83% adaptive calibration error, 3.0% of predictions flip when the option list is reversed, 35.6ms on CPU, with the failed use cases reported alongside the wins.

Open resource

Source: community catalog. Details and metrics reflect the published community snapshot and are not a performance endorsement.

Keep exploring

Similar to Verdict (Open-jev) launch thread

Browse all

Hacker News launch thread

Articles & threads

1,800-point thread debating whether typed decisions replace LLM calls for classification, routing, and scoring.

CMade byCommunity contributor

Launch thread by Diogo Almeida

Articles & threads

TypeSafe's founder on why RLCD-trained decision models are a shorter path to value than chat models.

CMade byCommunity contributor

Kalshi prediction-market bot

Articles & threads

Jev trades 15-minute and 1-hour BTC, ETH, and SOL markets on Kalshi.

CMade byCommunity contributor

jev-skill

Other lists

An awesome collection of Jev use cases, workflows, and agent skills.

WMade bywuyoscar471

awesome-jev (OmniJev)

Other lists

Papers, open reproductions and independent evaluations behind System One models and Jev.

OMade byOmniJev286

awesome-jev (kraayenjon)

Other lists

A curated list of Jev use cases, projects, SDKs, and resources. Jev is TypeSafe AI's System One model for fast, typed decisions in software — Choice, Score, and Noul with calibrated probabilities.

KMade bykraayenjon130

Introducing System One Models and Jev

Official

Launch post: architecture, RLCD training, pricing, Doom and Wikiracing demos, FAQ.

TMade byTypeSafe AI

Manifesto

Official

The case for machine-native intelligence built for software, not conversation.

TMade byTypeSafe AI

TypeSafe AI

Official

Company homepage, waitlist, and product overview.

TMade byTypeSafe AI