All decision model projects

Twin-stick shooter where Jev controls the enemies, making each move and shot as a typed decision.

Small browser game built to test how Jev handles unknown or unexpected input.

Browser 1v1 shooter in the style of Krunker where Jev decides move, aim, aim-down-sights, fire and jump about nine times a second.

Hosted site of playable board games in which Jev makes the computer player's decisions.

Web tool that analyses Telegram channel JSON exports locally for intent, quality, sentiment and speaker tone with Jev.

Independent early-access trial of Jev that measures its calibrated judgments and reports cost per decision.

Kiln blog post on optimizing only the harness around Jev for Chinese polyphonic-character reading, halving errors at a seventh of the cost.

LangChain post comparing Jev with LLM judges for agent evaluation on accuracy, repeatability, latency and cost.

OpenRouter benchmark of Jev 1.13 vs Claude Opus 5 on Banking77: 81.0% vs 84.4% accuracy at a fraction of the latency and cost.

Veris AI blog post where Jev decides when a caller has finished speaking, cutting agent interruptions from 52% to 11% of turns.

Independent experiments on Jev for search stopping, memory reranking and multi-hop relation selection, with stated limits.

Study comparing one direct Jev question per row with 12–14 Jev-scored dimensions and fitted weights on three classification tasks.

SREGym blog post on Kubernetes incident diagnosis with no LLM agent: Jev picks a root cause from collected evidence.

Experiment site testing with data whether Jev's judgment about a person changes when their profile changes.

Blog post testing Jev in 16,000 calls against two OpenAI models, covering where it wins, where it breaks and how to set thresholds.

Independent evaluation by Senko Rasic comparing Jev zero-shot on Croatian text against a self-fine-tuned BERT.

Research note testing whether Jev's typed answers can be manipulated by prompt injection, across 14,495 logged calls with raw logs.

Classmethod DevelopersIO independent test of Jev for LLM model routing, reporting 40 successful calls and 10 of 10 correct classifications.

Article cross-checking 108 statements to argue that Jev's calibration is not unique, including the author's own corrections.

Flavio Copes' hands-on walkthrough of Jev and the System One idea, aimed at developers who have only seen LLM marketing.

Practical guide to Jev covering the playground, Python and JS SDKs, raw HTTP calls and the agent skill.

Latent Space AINews launch-day roundup on Jev as a System One model that only decides.

Tutorial contrasting one-pass option probabilities from a decision model with an LLM that needed 11 passes to emit the same JSON.

LangChain blog guide on wiring Jev into an agent harness as a dangerous-action classifier and model router.

Hugging Face post on using a decision model to clear political fundraising emails out of a real inbox.

cellcog.ai page tracking Jev that keeps layered, dated records of what is claimed, conceded, measured and found.

News that Cloudflare's Clef-omni decision model takes text, images, audio and video in one request, with a Clef-flash price cut.

Curator note: TypeSafe's llms.txt revealed 18 official Jev cookbook recipes with runnable code, datasets and measured figures.

elvex blog on wiring Jev into an agent harness as a tool for search, approvals and context, citing 2,000 expense reports in 20 seconds.

Head-to-head eval page where TrustedRouter's gev-1.0 outperforms Jev on the same JevBench questions and also accepts images.