
GitHubInbox and support
jev-wechat-live
An abandoned WeChat overlay, published with the full post-mortem.
duckegg0623-create
5Updated Nov 13, 2026
855 projects, posts and guides about Jev, the System One model from TypeSafe AI. Each entry links to its source and shows the cost and speed its author reported.

Diogo Almeida
@CompleteSkeptic
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x cheaper (w/ output tokens free) • Frontier composable intelligence optimized for decisions AFAICT the shortest path to AI-based economic revolution

GitHubInbox and support
An abandoned WeChat overlay, published with the full post-mortem.
duckegg0623-create
5
GitHubOpen source
Make an existing SGLang or vLLM deployment answer as a decision service.
David-Lolly
6Guideyoutube.com/@AI-GPTWorkshop
Zubair Trabzada's AI Workshop: what Jev actually does, and why calling it an LLM misleads.

GitHubContext and memory
Context compaction for DeepSeek Harness, judged by Jev.
yangyu666
5
GitHubDocuments and OCR
A thousand arXiv papers, one Jev decision each, checked by an LLM judge.
stas4000
71,000 papers
GitHubAgents and browsers
Turn a prompt, or the code that runs one, into a Jev decision.
sumleo
6Guideyoutube.com/@Supabase
Supabase, in a short: Jev introduced in under a minute.


GitHubBenchmarks and evals
Research on which LLM-decision tasks Jev actually speeds up.
Tongyun1
4
Guideyoutube.com/@samwitteveenai
Sam Witteveen goes through the open decision models released in Jev's first week.

GitHubDocuments and OCR
Typed relation governance and safe archive plans.
anpicasso
3
GitHubBenchmarks and evals
Calibrated typed-decision models, one forward pass each.
Metask AI
1
Guideyoutube.com/@engineerprompt
Prompt Engineering puts Jev in the reranking step of a RAG pipeline: why vector search and cosine similarity fail, how a steerable reranker works, and how it compares with LLM rerankers and cross-encoders. A Colab notebook comes with it.

GitHubAgents and browsers
One ~0.4 s typed decision per DOM step, instead of an LLM turn.
ZephyrDeng
12~0.4 s
GitHubAgents and browsers
Type what you want, and the right menu item runs in any Mac app.
Rohit Mehta
5
GitHubOpen source
Typed decisions in the browser: one state, any number of options.
Nico Martin
41Guideyoutube.com/@JulianGoldieSEO
Julian Goldie’s hour-long course: the three question types, many decisions batched in one request, and ten use cases, from email sorting and lead scoring to internal linking across 586 pages and a browser agent finding flights.



GitHubOpen source
Knowledge graphs built from typed relation decisions, with the evidence kept.
chenmingtang830
40Guideyoutube.com/@KevBuildsApps
Kev makes Jev the decision maker inside HyperEdit, his open-source AI video editor, and walks through the whole setup with Claude Code. The repository is free on GitHub.

GitHubSDKs and integrations
Semantic decisions as ordinary Ruby: feels?, decide, score.
Qew7
20

GitHubSecurity and abuse
Screen an agent's tool calls with Jev before they run.
agent-chaperone
22Guideyoutube.com/@CoderOne
Coderone on Laya, the open System One model from Convai Innovations: what typed decisions are, why they replace a share of LLM calls, and Laya run locally to see if its figures hold up.
400MParameters
GitHubGames and real time
A 400M local model doing zero-shot decisions, and playing Doom.
Deepan Wadhwa


Guideyoutube.com/@daveebbelaar
Dave Ebbelaar in Python: a support-ticket classification first, then Choice, Score and Noul, several questions in one call, and latency and price next to Claude Haiku, Opus 5 and Fable 5.1.
7xSpeed-up
ArticleBenchmarks and evals
The loop replaced with a single Jev call, and the numbers published.
r6i.it
Alex Volkov
@altryne
This is actually insane. This uses @typesafeai Jev model, as a plugin in Claude to review all the un-nesseasary tool calls, and it takes 1s to run! Like, literally, 1 second to take my Claude session from nearly 1M to ... 86K tokens! 😮 Ask your claude to install it and be amazed Use this prompt ``` Install, and configure : https://github.com/tamaratran/fast-jev-compaction ```