Jev
TypeSafe's System One model. Typed decisions, no chat. Trained with RLCD. Output tokens free.
Feature a build or sponsor the directory. DM @pranavmore69
TypeSafe Jev · 15 September 2026
TypeSafe left stealth on 15 September 2026 with Jev. 13 first-party ships we cataloged, each with the original post.
Diogo Almeida, who worked on InstructGPT at OpenAI, trained Jev with RLCD, Reinforcement Learning for Calibrated Decisions. The model cannot write a sentence. It scores choices. Input is $0.042 per million tokens. Output is free. That is the whole pitch, and people started wiring it into crawlers, code review, and trading bots the same week.
Official note: Introducing System One Models and Jev. Waitlist: typesafe.ai. Announcement: the original post.
TypeSafe's System One model. Typed decisions, no chat. Trained with RLCD. Output tokens free.
Chrome extension that listens to YouTube audio, detects sponsor segments, and skips them in real time. Open source, BYOK.
Trading bot with Jev. Buy or sell from a price feed, then places orders on Monad via Kuru every 300ms.
Give it a company homepage. Jev finds Careers, scores roles against your profile. Job crawler rebuilt from 5 minutes to 20 seconds.
Simple Jev review bot. Shipped in the first hours of Jev access.
Open-source voice browser built on TypeSafe Jev.
Rust SDK for Jev. Typed async choices, scores, probabilities, and confidence.
Production layer for Jev: policy, confidence gate, shadow mode, recipes, eval CLI.
Wires Jev into Claude Code so every review finding is scored before you read it.
Proof-of-concept code vulnerability scanner using Jev. Flags risky code without an LLM writeup.
Playable Jev demo. A needle that panics in realtime. Possibilities, not text.
Turns AGENTS.md preferences into Jev-shaped checks. An intelligent linter for your agent.
Pi extension powered by Jev that hides low-value history from model context without deleting the session.
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 Show more
Just trying out Jev, I made a Chrome extension that: - Listens to your YouTube audio (optional) - Detects if it gets to a sponsor segment - Skips it ➡️➡️➡️ - All in real-time while costing ~$0.005 per video Prototype project, BYOK, open-source: github.com/trungdq88/yout…
I built a trading bot with Jev! Jev decides if it should "buy" or "sell", given the price feed of an asset pair, and executes real trades. It uses Monad to place the orders on Kuru's on-chain order book in every 300ms block. Demo link → jev-trader.vercel.app
I open-sourced the skill from this experiment: jev-job-hunter. Give it a company's official homepage. Jev finds the Careers entry point, identifies job pages, and scores each role against your profile. If you have a Jev API key, give it a run: github.com/hqman/JevScout
I rebuilt my job crawler with Jev. The task: start at a company's official homepage, find Careers, and identify jobs that match my profile. Before, with an LLM: ~5 minutes. After, with Jev: just over 20 seconds in my test. Every company organizes its website differently. Jev
so far, 3 very simple review bot, shipped within the first few hours of gaining access - github.com/devagrawal09/j… a more general coding assistant using only jev that i'm still working on - github.com/devagrawal09/j… another internal project at work that combines deterministic ast Show more
I've open sourced this if anyone wants to check it out github.com/moritzkremb/je…
Jev is built for decisions, not chat, so I made a Rust SDK for it 🦀 jev-rs gives Rust apps a typed, async-first way to get choices, scores, probabilities, and confidence from Jev. Would love to hear what Rust developers think 👇 github.com/abeldzan/jev-rs
TypeSafe Jev is great at structured decisions. I open-sourced the missing production layer: jev-harness → policy → confidence gate → shadow mode → recipes → eval CLI github.com/AntonioCoppe/j…
I wired TypeSafe's Jev into Claude Code so every code-review finding gets scored before I ever read it. First real run, I pointed it at the plugin's own file reader. It found a genuine bug I had shipped 20 minutes earlier. MIT: github.com/buchmark/claud…
Spent this morning with Typesafe’s Jev - a model that returns possibilities, not text. I made a needle that panics in realtime ask-jev.vercel.app
I know I'm late to the party. But here is my mandatory Jev project. It turns your preferences in AGENTS.md file into something Jev-shaped, so your ai agent can use it sort of like an intelligent linter? github.com/doeixd/jev-pre… @typesafeai
Built pi-jev-context: a Pi extension powered by TypeSafe Jev that hides low-value history from model context without deleting your session. Opt-in, reversible, and configurable. Keep useful context. Drop the noise. github.com/kevinpita/pi-j…