↑ Rising

State to typed decisions for software

Jev provides typed probabilistic decisions for software by processing state and questions in a single pass.

99 builds · 98 builders · First seen Sep 19 · 36.1× momentum

Your agent's guardrail is a frontier LLM judging every action.

Your agent's guardrail is a frontier LLM judging every action. TypeSafe's Jev — probabilities, not prose — ran Vercel's safety classifier 5–18x faster, and can't hallucinate by construction. https://www.ideabosque.com/l

IdeaBosque@ideabosque

4 views on XJev score 54

Jev Decision Graph

Jev Decision Graph transforms noul, choice, and score inputs into auditable multi-step workflows with typed nodes, confidence fallbacks, and traces.

Shay@Shay145287

2 views on XJev score 57

Jev Decision Engine

Jev is a pure decision engine that evaluates predefined decisions in parallel, offering a faster and cheaper alternative to traditional LLMs for bounded answer spaces.

Naise AI@naise_ai

21 views on XJev score 61

Jev: A System One Model for AI Decisions

Jev is a new category of AI model designed for decision-making rather than generation, offering low latency and cost-effective inference through three primitives: Choice, Score, and Noul, all with calibrated confidence scores.

parth jain@parthjain_1

113 views on XJev score 69

Jev Decision Model Performance Comparison

Jev is a decision model that provides choices, probabilities, and confidence scores, making it suitable for tasks like chatbot intent routing, and it was tested against several LLMs for accuracy, speed, and cost.

Rouzbeh@JustRouzbeh

1.5K views on XJev score 59

Play video

Jev X Scorer

Jev is a model that selects the best option from a given set and provides a confidence score, used here in a browser extension to grade X drafts against a ranking algorithm.

Build Fast with AI@BuildFastWithAI

9 views on XJev score 56

Jevable Showcase

Jevable.com is a live showcase of 194 projects built with Jev, a judgment layer for software that takes a state and a question to return a Choice, Score, or Probability with confidence.

Priya@Priyannkaaaa

41 views on XJev score 61

Jev AI Decision Model

Jev AI is a 'System One Model' designed for fast, structured decisions within software, functioning as a sorter rather than a writer by choosing from predefined answers based on context and a narrowly defined question.

Good AI@GoodUseOfAI

37 views on XJev score 54

Jev explained in 1 minute 👇 Jev lets you turn an LLM into a structured evaluator.

Instead of asking it to “classify” something, you define the possible outcomes + criteria, and Jev evaluates the input against them.

Pranav Ahuja@pranavahuja270

13 views on XJev score 48

Jev Decision-Only AI Model

Jev is an AI model that provides typed answers with probabilities for closed questions, functioning as a decision-only system.

0xHades@OnlyHades_

85 views on XJev score 53

Jev by TypeSafe AI

Jev is a System One model designed for structured judgments, accepting context and a scoped question to provide program-ready answers with probabilities, including outputs like Choice, Noul, and Score, and supporting runtime natural language labels without retraining.

Meng Li@dlimeng192048

34 views on XJev score 52

Jev Decision Model Performance

This project tested the Jev decision model's speed against other LLMs, finding it to be significantly faster.

Maxshot AI@MaxshotAI

43 views on XJev score 33

Jev: Probabilistic Decision Model

Jev is a model that provides probabilistic decisions on defined questions instead of generating text, operating in a single parallel pass with typed inputs and outputs.

Robot-man@RobotsTJ500

202 views on XJev score 75

Jev: Type-Safe Probabilistic Reasoning

Jev is a non-chat model that takes typed state and questions, returning calibrated probabilities in a single pass without generating text.

Hazem Omier@hazemomier

25 views on XJev score 64

Play video

JevPick

JevPick is a tool that uses the Jev model to analyze a prompt's requirements and recommend the cheapest capable AI model for handling it, rather than defaulting to the most expensive option.

Abinashi singh@singh_abinashi

82 views on XJev score 61

Jev: Bounded Question Answering

Jev is an AI system that answers bounded questions with a probability, a selection from a list, or a score, operating in a single pass with low latency and cost.

Sarcastic Badger@Entropy_Badger

Posted 12h agoJev score 68

Jev: Typed Decisions for Bounded Answer Spaces

Jev is a system that takes unstructured input and outputs typed decisions with probabilities, designed for narrow, bounded decision-making tasks rather than generative language.

Carpe Diem@Ask_CarpeDiem

136 views on XJev score 58

Jev AI: Agent Decision Layer

Jev AI provides a cost-effective mechanism for agents to determine when not to act, returning structured decisions and confidence levels for routing, risk scoring, approval, or human escalation.

Shinka - AI@ShinkaIoT

307 views on XJev score 54

Interlinked @muse & Jev 193.6x faster, 444.6x cheaper.

Interlinked @muse & Jev 193.6x faster, 444.6x cheaper. @typesafeai's in house math for Jev against frontier LLMs on decision workloads. Typed answers replace token streams. Muse runs the workflow. Jev answers in ~

DARWIN@D4RW1NEXE

17 views on XJev score 52

Play video

Jev Model for Structured Output

Jev is a fast model from TypeSafe AI designed for structured output tasks like classification and routing, capable of answering in under a second.

Syphax Ait oubelli@syphhhhhhhhhh

9 views on XJev score 37

Jev: System One Decision Model

Jev is a 'System One' model that bypasses text processing to deliver rapid, structured decisions for tasks like ticket triage and model routing, operating autonomously above 90% confidence.

guerrechon chatelain@Nohcreug

6 views on XJev score 61

Jev Prediction Market Synthesis Engine

Jev is a probabilistic scoring engine designed for decision synthesis in event contracts, evaluating Yes/No probabilities, Choice options, and Scoring.

Mo@morobsharif

9 views on XJev score 42

Play video

JEV for Construction Data Analysis

JEV is a small, fast model designed to answer structured questions about data, such as yes/no, scores, or selections from a list, making it efficient for analyzing project data like meeting transcripts.

Kenny Shultz@PermitZIPhvac

19 views on XJev score 53

Jev: A System One Model for Efficient AI Evaluation

Jev is a System One model that takes state as input and outputs typed choices, scores, or booleans with calibrated probabilities, avoiding string generation and performing parallel evaluations.

Ayyaz@ayyazdev

50 views on XJev score 69

TypeSafe Jev Agent Stack Integration

This project integrates TypeSafe Jev into an agent stack, enabling the system to return typed decisions like probabilities, choices, and scores, offering speed advantages over LLMs for numerical tasks.

Pierpaolo Villano@Pier135

11 views on XJev score 45

Jev Decision Engine

Jev acts as an intermediary for AI agents, making a small decision before each action to either execute directly, use a stronger model, ask the user, or block, providing only the decision and confidence level.

Venus Kalra@venuskalra

5 views on XJev score 61

LLM vs. Jev Decision Making

This project explains how Large Language Models (LLMs) answer questions, contrasts this with TypeSafe's Jev model, and demonstrates a use case for invoice triage with confidence gating.

Sumit@sumitp01

47 views on XJev score 48

$JEV

A public, free entry point for TypeSafe AI's Jev model, which takes user-defined states and typed questions to return structured, parallel answers without generating long text, suited for immediate program branching.

Jeff@cfm_sol

4.5K views on XJev score 60

Jev Decision AI

Jev is an AI model that analyzes text prompts and provides decisions in one of three formats: Yes/No with confidence percentage, selection from a list, or a score on a scale.

Josh with an eye@technogeek007

25 views on XJev score 57

Jev Independent Index

Jev is an AI model used to make fast, inexpensive yes/no or multiple-choice decisions, acting as a preliminary filter before more complex models handle ambiguous requests, significantly reducing costs with a minor accuracy trade-off.

Todd Dailey@twid

63 views on XJev score 68

Play video

Jev Model Quality Benchmark

This project evaluates the accuracy of the Jev model in classifying and routing simulated emergency dispatch incidents, testing its ability to identify incident types, assess urgency, select response units, and make disposition decisions.

Asta@asta_td

48 views on XJev score 60

Jev: Typesafe AI Decisioning

Jev provides typesafe AI responses, outputting structured data like Choices, Scores, or Booleans with probabilities, eliminating the need for string parsing.

Jev Watch@jevwatch

66 views on XJev score 61

Jev: AI Decision-Making

Jev is an AI system designed for rapid, cost-effective decision-making in software applications, functioning like a multiple-choice AI rather than an essay-writing one.

Harshal Suralkar@hvrshvl

14 views on XJev score 53

JEV: Structured Output for AI Decisions

JEV is an AI system that generates structured, typed outputs directly, eliminating the need for parsing or retries by bounding the output space before inference and accepting predefined answer types.

Jason Goldberg@betashop

279 views on XJev score 63

Jev AI Model Testing

Jev is an AI model that provides typed decisions for code to act upon, achieving 20/20 accuracy in testing.

pybankers@pybankers

15 views on XJev score 53

Jev: Typed Probabilistic Decisions

Jev is an AI system designed for machines that returns typed probabilistic decisions with a fast response time.

Nader Khaled@nrkhaled11

2 views on XJev score 54

Jev: Machine Native AI Model

Jev is a machine-native AI model that provides typed probabilistic decisions for software systems, utilizing Choice Score and Noul question primitives for structured, probability-based responses.

Apollo@ApollonVisual

40 views on XJev score 64

Jev: Typed Probabilistic Decisions

Jev is a model that provides typed probabilistic decisions instead of generating text, designed for routing, scoring, and real-time agents.

ShowTimeZ@_Show_TimeZ_

115 views on XJev score 74

Jev AI Model

Jev is an AI model that indicates its confidence level in its responses, unlike traditional AI that may present uncertain answers with false certainty.

Michael Kaiser@MichaelKaiserAI

59 views on XJev score 44

Typesafe AI for Decision Making

Typesafe AI is a system for structured decision-making problems that selects from predefined answers and provides probabilities, rather than generating text like a traditional LLM.

Nav@NavDoesTech

2.4K views on XJev score 55

TypeSafe AI System One Model

TypeSafe AI's System One Model, named Jev, processes messy input and predefined questions to produce structured, type-safe judgments with calibrated confidence, avoiding prose generation.

JimiLonbo@jimmy_longbow_

168 views on XJev score 56

Jev Typed Question Answering Model

Jev is a novel model that answers typed questions with associated probabilities, operating in approximately half a second.

Mihai Dragomirescu@mihai_respira

34 views on XJev score 52

TypeSafe AI Jev Model Integration

This project integrates the TypeSafe AI Jev model into a system layer to process unstructured state into typed decisions, aiming to eliminate issues with parsing unstructured data.

Taseed@TaseedMustafa

85 views on XJev score 51

Jev as an LLM Guardrail and Verification Layer

Jev acts as a verification layer for generative LLM responses, checking for safety, factual alignment, and policy compliance before the response is delivered to the user.

Kayo Digital@kayodigitalza

21 views on XJev score 50

Jev AI Model

Jev is an AI model that takes 'state' and typed questions to return typed decisions with calibrated confidence, excelling at classification tasks.

Ivan Khokhlov@ivantinkers

12 views on XJev score 47

Jev: Structured Decision-Making AI

Jev is an AI model that provides structured decisions, such as classifications, scores, or probabilities, instead of generating explanatory text, to help applications process information efficiently.

FoundOnArdy@FoundOnArdy

185 views on XJev score 52

Play video

Typesafe AI's Jev Model

Jev is an AI model that provides typed decisions, such as yes/no probabilities, list selections, or scores, instead of conversational text, operating significantly faster and cheaper than full LLM calls.

Oliver Prompts@oliviscusAI

563 views on XJev score 68

Jev: AI for Decisions

Jev is an AI model designed to make specific, bounded decisions within software applications, outputting typed decisions and probabilities rather than conversational text.

Nikhil Mourya@GonnabeNikhil

464 views on XJev score 56

Jev System One Model

Jev is a "System One" model designed for fast, structured decisions, accepting a state and typed questions to return structured answers with probabilities.

Srimugunthan Dhandapani@Srimugunthan_D

39 views on XJev score 61

Jev: A Decision-Making AI Model

Jev is an AI model that provides typed decisions, probabilities, and confidence scores for software to act upon, differing from traditional LLMs by not generating answers token by token.

QFS17@riabcevv

161 views on XJev score 67

Jev System One Model

Jev is a new System One model from TypeSafe AI that answers typed questions by returning a value and a confidence number for code branching.

Kev — Building with AI@alltechkevin

31 views on XJev score 42

Typesafe AI: Machine-Native Intelligence

Typesafe AI is developing machine-native intelligence that makes decisions within software, contrasting with previous chatbot approaches that operated alongside workflows. The project is currently testing its first system, JEV, focusing on decision latency and performance within local agent runtimes.

MotionViz@Motion_Viz

632 views on XJev score 43

Jev Performance Comparison

This project compares the performance and cost of Jev against existing LLM-based structured decisions, showing Jev to be significantly faster and cheaper.

AI Adjuncts@aiadjuncts

27 views on XJev score 46

TypesafeAI Jev Classifier Integration

This project integrates TypesafeAI Jev into a classifier to improve model, agent, and use case selection, resulting in significant cost, time, and evaluation improvements.

Root Access@root4ccess

19 views on XJev score 55

Jev Transaction Categorization

Jev is a tool that performs transaction categorization, achieving comparable accuracy to GPT-5.6 while being significantly faster and cheaper, and it can identify when it is not 100% certain about a categorization above a 0.85 confidence threshold.

AA Li@alilibx

56 views on XJev score 55

TypeSafe AI Jev Early Access

This project is testing Jev, a system for structured, typed decisions and confidence scores across classification, routing, scoring, and verification.

Shoh G Man@topshoh

49 views on XJev score 26

Jev: AI for Software Decisions

Jev is an AI model designed to make decisions within software applications by processing state and narrow questions to return typed answers with probabilities, enabling semantic if-statements where code handles the final branching and actions.

Abhijeet@abhijeetdevv

39 views on XJev score 61

Play video

Jev Model for Help Center Classification

This project implements the Jev model for real-time help center classification, analyzing user questions to identify the most relevant support article.

Valentin Sattinger@v_sattinger

23 views on XJev score 50

Jev Routing Function Testing

Jev was tested on core routing functions, demonstrating accuracy comparable to or exceeding current LLMs, with significantly improved speed and reduced cost.

Gus Fraser@gusfraser

493 views on XJev score 59

Jev Model Performance Evaluation

This project evaluates Jev, a probability-based model, on two real-world tasks: analyzing SEC filings and checking extracted facts, highlighting its strengths in narrow, yes-or-no questions and its limitations with complex document interpretation.

Shaun Regenbaum@RegenbaumShaun

49K views on XJev score 66

Jev Model for Legal Use Cases

This project tests the Jev model from typesafeai for legal applications within stll_app, noting that Jev is optimized for decisions rather than text generation and provides typed answers with probabilities.

Jan Kubica@jan__kubica

8.2K views on XJev score 45

Jev Independent Index

Jev is an AI model effective for classifying inputs into predefined categories when the criteria are explicitly defined within the input data, offering significant cost and latency advantages over existing models.

drew dillon@drewdil

27K views on XJev score 66

Typesafe AI Jev

Jev is an AI technology that provides faster initial question classification compared to traditional LLM calls, as demonstrated by Barnacle Labs.

Duncan Anderson 🍃🔋⚡️🤖@duncsand

61 views on XJev score 41

Jev Independent Index

Jev is an AI model that returns a choice or score with confidence based on predefined possible answers, rather than acting as a chat model.

Kalash Killy@MichalKiliany

31 views on XJev score 50

Jev: Typed AI Agent Decisions

Jev is an AI model that takes shared state and typed questions, providing direct code-usable answers with choices, scores, and probabilities, enabling efficient decision-making for AI agents.

adidshaft@adidshaft

194 views on XJev score 74

Play video

Jev: Fast Decision-Making for Agentic Systems

Jev is a new System One model from TypeSafe AI designed to accelerate decision-making in agentic systems by processing state and typed questions to produce calibrated decisions in under a second, eliminating the need for text parsing.

Elias Andualem@eandualem

947 views on XJev score 63

Jev Production Deployment

Jev has been deployed in production for Proofpress via Vercel AI Gateway, enabling default evidence review due to its speed and low cost.

Richard Tang@richardt830

75 views on XJev score 59

Play video

Jev Integration for AI Agents

This project integrates Jev by @typesafeai into a SaaS AI Agent to provide faster and more cost-effective responses compared to traditional LLMs, with a mechanism to identify when a full LLM is still required.

Lennart@LennartPrange

38 views on XJev score 49

Play video

Jev: A System One Model for Software Decisions

Jev is a "System One" model from TypeSafe AI designed for making fast decisions within software by evaluating predefined choices in parallel and returning typed values with probability distributions, rather than generating text token by token.

Jurly@jurlycat

1.8K views on XJev score 75

Jev: A TypeScript-Ready Decision Model

Jev is a model that takes a state and typed questions to provide choices, scores, and yes/no probabilities for code to branch on, enabling automated decision-making in applications.

Mark Argyelan@markargyelan_

44 views on XJev score 56

Jev: An Independent Index for Decision Making

Jev is a model that takes program state and typed questions as input, returning calibrated yes/no answers, choices, or scores with low latency, designed to handle decision-making tasks separate from text generation.

Alex Freitas@AlexFreitasAI

137 views on XJev score 67

Play video

TypeSafe AI System One (Jev) vs. RLHF LLMs

This project compares TypeSafe AI's System One model, Jev, against current RLHF-based LLMs for decision-making tasks, specifically on a smart model routing use case.

Kishan Lal@kishan0725

987 views on XJev score 34

TypeSafe Jev Model

The Jev model is an AI component that makes structured decisions for immediate program use, rather than engaging in dialogue or lengthy explanations.

21Borov@21_borv

Posted 1d agoJev score 56

Jev: An Independent Index for AI Engineering

Jev is a framework designed to optimize AI agent workflows by identifying and automating decisions that do not require lengthy LLM responses, thereby building reliable loops around them.

Avid@Av1dlive

68 views on XJev score 57

Jev Decision Model Performance

Jev is a decision model that demonstrated superior speed and efficiency compared to other LLMs in evaluations.

OpenRouter@OpenRouter

17K views on XJev score 50

Jev Primer

This project provides an objective explanation and a clear demonstration of Jev, a new model from @typesafeai that answers questions with a fixed shape based on a given situation, enabling its use in code.

Marc Bara@marc_bara

28 views on XJev score 42

Jev Use Case Expansion

This project explores and integrates 18 use cases for Jev within a personal bot harness, demonstrating significant improvements in model efficiency and decision-making.

Divine 〽️achine@Divine_machine

45 views on XJev score 67

Jev: A Decision Classifier

Jev is a decision classifier that provides confidence scores for choices, rates inputs against ordered levels, or answers yes/no questions, enabling parallel processing of multiple queries unlike sequential token generation.

Arya Chandra@AryacChandra

448 views on XJev score 45

Play video

Jev Model Testing

This project tests Jev, a new model from TypeSafeAI, which uses Reinforcement Learning for Calibrated Decisions (RLCD) to interpret commands and return probabilities, demonstrating faster and cheaper performance on structured workflows.

Kevin Ly@kevinlysocial

60 views on XJev score 56

Jev: Fast, Calibrated Decisions

Jev is an AI system designed for fast, System 1-style decisions by selecting from a fixed list of answers with associated probabilities, unlike slower, System 2 LLMs that generate text.

Bala Venkatesh@balavenkatesh22

261 views on XJev score 54

Jev: AI Decision Engine

Jev is an AI system that makes typed decisions and probabilities based on state, contrasting with traditional AI models that generate text.

Aakash@aakash_env

67 views on XJev score 67

Building a Harness with Jev

Jev is a 'System One model' from TypeSafe AI that handles decisions by returning typed answers with probabilities based on state and questions, rather than generating text.

cv usk@cv_usk

94 views on XJev score 61

Jev: AI Agent Decision Orchestration

Jev acts as a decision-making layer for AI agents, choosing the next move by splitting intelligence from execution and enabling parallel processing of decisions.

Misha_Cripto@MishaAi333

83 views on XJev score 73

Jev Decision Model for Y Combinator Companies

This project utilizes Jev, a decision model from @typesafe, to evaluate all Y Combinator companies, returning probabilities rather than text.

Rohan Bhanot@RohanBhanotAI

39 views on XJev score 57

Jev: Non-Autoregressive Decision Model

Jev is a non-autoregressive decision model that processes application state and typed questions in a single forward pass, returning calibrated probabilities for efficient filtering, gating, and selection tasks.

Hemant@heman10x

449 views on XJev score 69

Jev Decision Engine

Jev is a system designed for making rapid, structured decisions by processing state inputs and answering typed questions with numerical outputs, enabling automated software actions without human-readable text generation.

amr@AmrTawfik160

98 views on XJev score 62

Jev: Fast, Accurate Decision Model

Jev is a new model that generates zero text, taking messy input and providing a clean answer with a confidence score, significantly faster and cheaper than traditional LLMs for tasks like simple decision-making.

Ling Li@LingLiAI

18 views on XJev score 52

Play video

Jev: Decision Intelligence Model

Jev is a decision-making AI model that provides typed answers with probabilities and confidence scores, trained using RLCD for reliable probability interpretation, and is significantly faster and cheaper than traditional LLMs for specific tasks.

Alcides Ticlla@AlcidesTicllaCh

46 views on XJev score 73

Jev: A System One Model for Classification and Scoring

Jev is a "System One" model that returns classifications, scores, and probabilities, unlike text-generating LLMs, and can evaluate multiple dimensions of AI agent sessions in parallel at a lower cost.

Ameya Kanitkar@aktwits

72 views on XJev score 63

Jev (TypeSafe) Model

Jev (TypeSafe) is a System One model designed for classification and routing tasks, providing calibrated decisions instead of text to improve speed and reduce costs.

Himanshu Kumar@mehkra1

19 views on XJev score 43

Jev: AI Judgement Model

Jev is a specialized AI model designed for performing judgements and verification tasks, acting as a quality control inspector for AI agents and workflows, rather than a general-purpose code or chat model.

sunshy@EthanChen213

16 views on XJev score 52

Jev AI for Customer Feedback Analysis

Jev is an AI tool designed to classify customer feedback into categories like bugs or feature requests and identify duplicate issues, returning typed answers with calibrated probabilities instead of generating text.

Chris DiNicolas@ChrisDiNicolas

3 views on XJev score 69

Jev AI for Customer Feedback Analysis

Jev is an AI model designed to classify customer feedback into categories like bugs or feature requests and identify duplicate issues, offering a cheaper, faster, and more accurate alternative to traditional LLMs for these specific tasks.

Chris DiNicolas@ChrisDiNicolas

58 views on XJev score 69

TypeSafe AI Jev System One Model

TypeSafe AI's Jev is a System One Model that takes context/state as input and returns fast, typed probabilistic decisions for software to act on, rather than generating strings like traditional LLMs.

Cliff Baldridge 😎 #AWE #AWE2026 #AR #VR #XR #AI@cliffbaldridge

79 views on XJev score 44

TypesafeAI Jev for Assistant Routing

This project utilizes TypesafeAI's Jev mechanism to improve the routing accuracy and speed of a family assistant's tests across multiple languages, significantly outperforming keyword matching.

Ahmed Nadar@ahmednadar

125 views on XJev score 68

Play video

AI-Powered Puzzle Game Bot

A bot plays a puzzle game by querying an AI model for tap probabilities before each move, utilizing live API calls for decision-making.

Michaël Ménard@mikemenard_com

144 views on XJev score 72

Play video

Live Sales Call Analyzer

This project uses Jev, an AI model that makes decisions and classifies things, to analyze live sales calls by dissecting conversations, identifying key deal signals, and triggering actions to assist sellers.

Mike Fishbein@mfishbein

732 views on XJev score 60

Jev as a Typed Judgment Layer

This project provides a practical guide on implementing Jev as a typed judgment layer in production, detailing how to design questions, manage uncertainty, and test workflows.

Jun Penn@heyjunpenn

876 views on XJev score 54

Jev: A Decision-Making AI Layer

Jev is an AI model that provides typed answers with probabilities to questions about a given state, functioning as a 'smart if statement' for automated decision-making.

Wanderer@invinci

77 views on XJev score 66

TypeSafe AI Jev Decision Engine

Jev is an AI decision engine that evaluates typed schemas directly in a single forward pass, bypassing the text generation of LLMs for faster and more reliable routing choices.

Nox|@noxflux

47 views on XJev score 66