What people are shipping with Jev.

built using jev @typesafeai https://t.co/CaKowqwRqW

Inspired by this research (and priors), and, @GoodfireAI catches reward hacks in the activations, but closed coding agents don’t give you those. So I built the other half: structural denies on graders/hidden tests + a @t

Jev AI dropped. @typesafeai So I built my own. NALURI by MEJALISM CORP. More intelligence. More agency. More acceleration. Build the future before the future builds you. ⚡🤖 Ad video below. https://t.co/aM2ngDz992 https:

Jev is insane. SOUND ON!!! I made music just by talking to it. "Add a snare on the backbeat." "Make it darker." "Way faster." "Add scratch on the offbeats and make it trippier." Every sentence becomes a new beat in ~200m

Dumping 50+ tool schemas wastes 90% context. Tool search fixes bloat, but costs an extra LLM round-trip (+2s) and fails on 7B models. tool-prune does it pre-flight in 0.4ms. https://t.co/BCAb9hvoPH Also, saw 100% on Berk

Jev od @typesafeai zmieni bardzo dużo, jeżeli chodzi o zastosowanie AI w firmach, jako że nie jest llmem, tylko modelem System One. Jeżeli ktoś chce się więcej dowiedzieć, jak ten model działa, odsyłam do ich docs: https

Introducing Bespoke Nimble: an open data, open model, open recipe for an open Jev. Code and info: https://t.co/rBzpX3KpHt Model: https://t.co/lpqAMT7wm8 Data: * A new data curation recipe called contrastive data curation

https://t.co/SAqnhhLUTr now finds the makers you should meet, near you 🤯 it's crazy accurate and built with Jev by @typesafeai my #1 match was @martbln_dev funnily enough we were DMing about meeting up yesterday 😂 find

TypeSafe raised $40M for Jev: typed decisions in one forward pass, calibrated. Closed weights, waitlist. I'm building the open version on one RTX 3090: Qwen3.8-27B, Apache-2.0, every number a committed run. No weights ye

Introducing Jev Reviewer, a tool designed for systematic reviewers to quickly find and extract the information from research articles. App: https://t.co/ScFHy5dYFR GitHub: https://t.co/gaTlohbGc1 https://t.co/jAKFWfQaJw

You can have Jev at home now! With live hotdog detection. Powered by Google's DiffusionGemma, the recipe runs on a single DGX Spark, _might_ also work with other Nvidia setups but untested (please let me know if it works

@typesafeai Jev is FREAKING SICK!! Playing around with it to create an experimental RLS Linter for @supabase. You can check it out here: https://t.co/DI796mSm3V or view the demo 📺 https://t.co/wKMGXSYJMk

👀 55 people right now exploring JEV use cases at https://t.co/z0izy4XNao @typesafeai https://t.co/41QGinihaF

Inspired by @typesafeai's Jev, I built sarvam-jev - an open engine that reads typed decisions straight out of an Indic model's logits instead of generating JSON token by token. Same sarvam-1 weights, 8.8× faster than mak

Time taken seems to be increasing over time, @typesafeai ' JEV seem to be getting busier 👀 Try https://t.co/V7CgA6iISc https://t.co/IbZapgMPVx

Interesting idea from @TypeSafeAI: Jev, a “System One Model” built for decisions, not text. Instead of generating JSON and parsing it, Jev returns typed decisions + probabilities. LLMs generate. Jev decides. https://t.co

JevPot. Jev, plus jackpot A weekend build for anyone chasing the dream of a big win — four number strategies scored together by Jev, ranked under 3 seconds, no sentence ever parsed. 6 requests. 33 judgments. Under 7,000

@thisiskp_ A PG CHECK constraint in plain english managed by Jev ;) https://t.co/UJ1eRWtaGm https://t.co/OoN1i1NPX6

Proof-of-concept code vulnerability scanner using typesafe's JEV model: https://t.co/mwhGkhpCfj Should be fun to try out on some CTFs. Currently only flags risky/vulnerable but could easily be extended to have an LLM-Age