# Jev Library

> An independent community directory of 719 open-source projects and 92 resources built on Jev, TypeSafe AI's System One model for typed, probabilistic decisions. Catalogue snapshot: 2026-09-21.

Jev takes program state plus a typed question and returns a typed decision — a Choice, a Score, or a Noul value — each with a probability distribution and a confidence. This directory catalogues what people have built with it: SDKs, agent tooling, browser and computer-use agents, applications, games, benchmarks, and writing.

This directory is independent and not affiliated with TypeSafe AI. Every entry links to its own public source; star counts and metrics reflect the snapshot date above and are not an endorsement.

## Start here

- [Jev Library home](https://jevlibrary.dev/): the full searchable directory.
- [Learn Jev](https://jevlibrary.dev/learn): how the three primitives work, plus 12 copy-ready starter projects.
- [Resources](https://jevlibrary.dev/resources): official documentation, articles, and community lists.
- [About and methodology](https://jevlibrary.dev/about): where entries come from and how they are reviewed.
- [Full catalogue](https://jevlibrary.dev/llms-full.txt): every entry, one line each.
- [New entries feed](https://jevlibrary.dev/feed.xml)

## Editor's picks

- [Jev Ultrafast](https://jevlibrary.dev/projects/browser-use-jev-ultrafast): A browser agent that turns a single goal into a sequence of lightning-fast actions.
- [Jev Trader](https://jevlibrary.dev/projects/jarrodwatts-jev-trader): One trading decision every Monad block. An experiment in real-time market intelligence.
- [Jev plays Mario](https://jevlibrary.dev/projects/fhshaik-typesafe-mario): Less talking. More jumping. Watch Jev navigate Super Mario Bros. from game state.
- [A home that understands](https://jevlibrary.dev/projects/abovecolin-ha-jev): Turn the state of your home into typed decisions, sensors, and smarter automations.
- [Mobile Jev](https://jevlibrary.dev/projects/droidrun-mobile-jev): An Android agent powered by Jev decisions, with a live studio to watch every step.
- [TypeSafe MCP](https://jevlibrary.dev/projects/itsmostafa-typesafe-mcp): Bring typed Jev judgments to your coding assistant through a single MCP server.

## Learn Jev — starter projects

- [Signal checker](https://jevlibrary.dev/learn/signal-checker): Paste a message and see how strongly it expresses one idea. The smallest useful Jev project. Teaches Noul, State design, API setup; about 15–20 min.
- [Inbox triage](https://jevlibrary.dev/learn/inbox-triage): Turn a customer message into a team, an urgency level, and an escalation signal in one call. Teaches Choice, Score, Noul, Parallel questions; about 30–45 min.
- [Confidence router](https://jevlibrary.dev/learn/decision-router): Use confidence to decide when software can proceed and when a person should take over. Teaches Confidence gates, Fallbacks, Observable decisions; about 45–60 min.
- [Feedback prism](https://jevlibrary.dev/learn/feedback-prism): Turn a pile of product feedback into themes, severity, sentiment, and concrete-request signals. Teaches Batching, Multi-axis analysis, Review queues; about 45–60 min.
- [Content preflight](https://jevlibrary.dev/learn/content-preflight): Check a draft’s hook, clarity, specificity, and call to action before it goes live. Teaches Composite scoring, Rubrics, Actionable UI; about 45–60 min.
- [Log first responder](https://jevlibrary.dev/learn/log-first-responder): Score noisy log events before sending only the useful ones to expensive analysis or a human. Teaches Triage pipeline, Cost control, Safe logging; about 60–90 min.
- [Skill caddy](https://jevlibrary.dev/learn/skill-caddy): Pick the one capability a coding agent needs now instead of loading every tool into context. Teaches Dynamic choices, Tool routing, Abstention; about 60–90 min.
- [Browser next-step lab](https://jevlibrary.dev/learn/browser-next-step): Let Jev choose among indexed page actions while a deterministic harness owns the browser. Teaches Action spaces, Observe/decide/act, Loop limits; about 90 min.
- [Command warden](https://jevlibrary.dev/learn/command-warden): Add a fail-closed judgment layer before an agent runs a shell command—without pretending it is a sandbox. Teaches Guardrails, Fail closed, Policy layering; about 60–90 min.
- [Snake instinct](https://jevlibrary.dev/learn/snake-instinct): Build a tiny game where code owns physics and Jev chooses one legal move per tick. Teaches Structured state, Legal actions, Replayability; about 60–90 min.
- [Home anticipator](https://jevlibrary.dev/learn/home-anticipator): Simulate a smart home that evaluates many possible needs at once, but changes nothing without approval. Teaches Fan-out, Contextual relevance, Human approval; about 60–90 min.
- [Calibration arcade](https://jevlibrary.dev/learn/calibration-arcade): Create a tiny labeled benchmark and learn whether confidence matches correctness—not just accuracy. Teaches Evals, Calibration, Selective risk; about 90–120 min.

## Recently added

- [NanoJev](https://jevlibrary.dev/projects/tianyucodings-nanojev): A nano replica of Jev: parallel decisions, dynamic candidates, and an end-to-end training pipeline.
- [jev-lint](https://jevlibrary.dev/projects/mizchi-jev-lint): lint text in code by jev scorerer
- [OpenDecision](https://jevlibrary.dev/projects/deepanwadhwa-opendecision): OpenDecision is an open-source semantic decision engine like typesafe's jev.
- [awesome-jev-use-cases](https://jevlibrary.dev/resources/walidboulanouar-awesome-jev-use-cases): Awesome list of TypeSafe AI Jev use cases: 74 demos ranked by likes, 150+ GitHub repos, limits, cost and API examples. CC0, sponsored by AY Automate.
- [jev-dataops](https://jevlibrary.dev/projects/renagao-jev-dataops): An open-source JEV-powered workbench for streaming data selection, quality evaluation, automatic LoRA training and held-out model evaluation.
- [awesome-jev (BeatAPI)](https://jevlibrary.dev/resources/beatapi-awesome-jev): We only curate source-reviewed JEV-related projects with 100+ GitHub stars — integrations, tools, open models, and experiments. Live gallery: beatapi.io/awesome-jev
- [RoboJEV](https://jevlibrary.dev/projects/lykycy123-robojev): Two-stage JEV control of a Franka Panda in MuJoCo
- [pi-jev-router (philippdubach)](https://jevlibrary.dev/projects/philippdubach-pi-jev-router): A minimal Pareto-optimal OpenRouter model router for pi, based on Jev
- [jevals](https://jevlibrary.dev/projects/openlayer-ai-jevals): Agent evals and guardrails in one request. Built on Jev, Kev and Laya.
- [Jev-Quantum](https://jevlibrary.dev/projects/karminski-jev-quantum): 亚微秒级 System-1 模型，准确率服从高斯分布
- [twitter-jev-guard](https://jevlibrary.dev/projects/qs-lll-twitter-jev-guard): 使用 TypeSafe Jev 在 X/Twitter 时间线上识别低质量、垃圾和广告帖子，并在文字区域显示醒目的半透明水印。
- [open-spark-jev](https://jevlibrary.dev/projects/abhishek085-open-spark-jev): Open-source, local decision models inspired by TypeSafe’s Jev and System One - built on Qwen3 for NVIDIA DGX Spark.
- [jev-leftpad](https://jevlibrary.dev/projects/f-jev-leftpad): Left-pad strings with TypeSafe AI's Jev. For reasons.
- [OpenJev (zhangcy122)](https://jevlibrary.dev/projects/zhangcy122-openjev): OpenJev: Open-source alternative to TypeSafe Jev. Typed probabilistic decision API (Choice, Noul, Score) powered by open LLMs & constrained logprob calibration.
- [awesome-jev-live](https://jevlibrary.dev/resources/wh000wh000-awesome-jev-live): Awesome Jev — evidence-graded index of TypeSafe System One: SDKs, MCP tools, agents, apps and open models. 20 languages, rebuilt every 2 hours.
- [JevLoop](https://jevlibrary.dev/projects/zjunlp-jevloop): The agent loop where decisions don't cost a large language model call. Zero deps, runs offline, no API key needed.
- [jevcache (kushals256)](https://jevlibrary.dev/projects/kushals256-jevcache): Skip expensive LLM calls when TypeSafe Jev says same intent. OpenAI-compatible local cache proxy — npx @kushalicious/jevcache
- [dohnuts](https://jevlibrary.dev/projects/psiace-dohnuts): Dohnuts builds small multimodal models for direct decisions. -> System One model
- [jev-minesweeper](https://jevlibrary.dev/projects/comoc-jev-minesweeper): TypeSafe Jev (System One) にブラウザ上のマインスイーパーを解かせるデモ
- [nanojev-arena](https://jevlibrary.dev/projects/caijinchun-nanojev-arena): NanoJev Snake Arena: 1v4 human-vs-AI battleship + 100-agent swarm simulator. Local demo of Jev System-One model (open-source mini replica).

## Official (19 entries)

Docs, SDKs, and resources from TypeSafe AI.

- [System One adapter (Python)](https://jevlibrary.dev/projects/typesafe-ai-system-one-adapter-python): Official drop-in TypeSafeClient replacement backed by OpenAI, Anthropic, and compatible LLM APIs, for comparing Jev against chat models.
- [TypeSafe JavaScript SDK](https://jevlibrary.dev/projects/typesafe-ai-typesafe-sdk-js): Official TypeScript/JavaScript client with inferred answer types. npm install @typesafe-ai/sdk.
- [TypeSafe agent skills](https://jevlibrary.dev/projects/typesafe-ai-skills): Official agent skill for Claude Code, Codex, and compatible agents: primitives, patterns, and how to structure evaluations.
- [TypeSafe Python SDK](https://jevlibrary.dev/projects/typesafe-ai-typesafe-sdk-python): Official sync and async Python client. pip install typesafe-sdk.
- [TypeSafe AI](https://jevlibrary.dev/resources/typesafe-ai): Company homepage, waitlist, and product overview.
- [Documentation](https://jevlibrary.dev/resources/documentation): Introduction, primitives, patterns, cookbooks, HTTP API, and SDK references.
- [Quick start](https://jevlibrary.dev/resources/quick-start): Shortest path from an API key to a typed decision.
- [Playground](https://jevlibrary.dev/resources/playground): Paste a state, add questions, and see typed answers in the browser.
- [HTTP API reference](https://jevlibrary.dev/resources/http-api-reference): Request and response contract for POST /v1/systemone.
- [Primitives](https://jevlibrary.dev/resources/primitives): Choice, Score, and Noul: the three question types and what they return.
- [Patterns](https://jevlibrary.dev/resources/patterns): Speculative fan-out, confidence-gated routing, composite scoring, and intent routing.
- [Cookbooks](https://jevlibrary.dev/resources/cookbooks): Reproducible recipes: parallel questions, reranking, guardrails, citation checks, extraction, hierarchical classification.
- [All 19 entries in Official](https://jevlibrary.dev/categories/official)

## SDKs & clients (61 entries)

Community clients for languages without an official SDK.

- [advocaat](https://jevlibrary.dev/projects/pithings-advocaat): A small, type-safe client for asking AI questions about your data, powered by TypeSafe Jev.
- [typesafe-sdk (joshmn)](https://jevlibrary.dev/projects/joshmn-typesafe-sdk): Ruby client for typesafe.ai
- [typesafe-ai](https://jevlibrary.dev/projects/twister915-typesafe-ai): Typed TypeSafe AI clients for Rust, with async and blocking backends and observable retries.
- [typesafe-dotnet-sdk](https://jevlibrary.dev/projects/saibimajdi-typesafeai-dotnet-sdk): Community .NET SDK for the TypeSafe AI System One API — typed noul, choice, and score questions with structured, confidence-scored answers. Not affiliated with TypeSafe AI.
- [TypeSafeAI.Net](https://jevlibrary.dev/projects/hawxy-typesafeai-net): .NET SDK for the TypeSafe AI platform
- [typesafe-go (cole-gillespie)](https://jevlibrary.dev/projects/cole-gillespie-typesafe-go): unofficial go SDK for typesafe AI, with typed answers, retries, and context support
- [typesafe_sdk](https://jevlibrary.dev/projects/nshkrdotcom-typesafe-sdk): An idiomatic, type-safe Elixir port of the official TypeScript AI SDK (ai / ai-sdk) providing unified LLM integrations, streaming text and structured outputs, tool calling, and agentic workflows. Jev is their current flagship model and is the first System One model.
- [jevgo](https://jevlibrary.dev/projects/fgn-jevgo): Go client for TypeSafe AI's System One API (Jev), with optional Langfuse instrumentation
- [jevclient](https://jevlibrary.dev/projects/abovecolin-jevclient): Async Python client for TypeSafe Jev. Typed questions in, probabilities and choices out, no prose to parse.
- [jod](https://jevlibrary.dev/projects/mateonunez-jod): Semantic schemas over TypeSafe's Jev — validate the state locally, then project typed answers.
- [jev-go (guillemus)](https://jevlibrary.dev/projects/guillemus-jev-go): Unofficial Go SDK for TypeSafe AI's Jev API
- [jev-go (Gaurav-Gosain)](https://jevlibrary.dev/projects/gaurav-gosain-jev-go): Go client for TypeSafe's System One API and its model Jev: typed judgments and calibrated probabilities instead of generated text
- [All 61 entries in SDKs & clients](https://jevlibrary.dev/categories/sdks)

## Integrations (33 entries)

Jev inside frameworks, gateways, and platforms.

- [eve](https://jevlibrary.dev/projects/vercel-eve): Vercel's open agent framework, which ships Jev as the default evaluation model in its experimental evaluate path.
- [ai-cli](https://jevlibrary.dev/projects/vercel-labs-ai-cli): Vercel Labs terminal CLI that can run Jev as the evaluation model for its evaluate command.
- [ruby_llm-typesafe](https://jevlibrary.dev/projects/kieranklaassen-ruby-llm-typesafe): TypeSafe structured-output provider for RubyLLM 2
- [A home that understands](https://jevlibrary.dev/projects/abovecolin-ha-jev): Turn the state of your home into typed decisions, sensors, and smarter automations.
- [neo4jev](https://jevlibrary.dev/projects/jexp-neo4jev): Typesafe.ai System One Model Jev navigating a Neo4j graph by using a classifier over neighbouring relationships
- [typesafe-jev-workflow](https://jevlibrary.dev/projects/giesn-typesafe-jev-workflow): Async LangGraph workflow that gets a typed Jev Choice (invoice or general) and routes each inbound email to the matching handler.
- [a0-typesafe-ai](https://jevlibrary.dev/projects/3clyp50-a0-typesafe-ai): TypeSafe AI Jev judgments for Agent Zero, with typed tools and probability cards.
- [typesafe-ai-rails](https://jevlibrary.dev/projects/genierobot-typesafe-ai-rails): Community Rails integration on the typesafe-sdk gem: configuration, persisted usage and cost telemetry, and opt-in confidence policies.
- [pg_typesafe](https://jevlibrary.dev/projects/giuliosmall-pg-typesafe): Pre-alpha PostgreSQL extension for TypeSafe AI (Jev) categorical classification
- [laravel-typesafe-jev](https://jevlibrary.dev/projects/butochnikov-laravel-typesafe-jev): Unofficial Laravel integration for TypeSafe Jev AI with typed responses, async requests, scoped dependency injection, and testing fakes.
- [Jev4Mellea](https://jevlibrary.dev/projects/soundblaster-jev4mellea): Jev adapter to Mellea
- [judging-with-typesafe](https://jevlibrary.dev/projects/carlsonchik-judging-with-typesafe): Скилл для агентов Letta: суждения по критериям через TypeSafe System One (Jev)
- [All 33 entries in Integrations](https://jevlibrary.dev/categories/integrations)

## Agent tooling (168 entries)

Gates, routers, reviewers, MCP servers, and skills for coding agents.

- [jev-review (devagrawal09)](https://jevlibrary.dev/projects/devagrawal09-jev-review): A staged code-review workflow and local dashboard built with TypeSafe Jev.
- [foreman](https://jevlibrary.dev/projects/thruwire-foreman): Software Factory Foreman: an agent supervisor that uses Jev decisions to keep coding agents on task.
- [jev-review (NiazMorshed2007)](https://jevlibrary.dev/projects/niazmorshed2007-jev-review): Local-first MCP plugin for continuous software-quality review by AI coding agents, powered by Jev.
- [jev-router (gargpratyush)](https://jevlibrary.dev/projects/gargpratyush-jev-router): Route to the cheapest model in claude code for your task using jev-router
- [jev-mcp (jkudish)](https://jevlibrary.dev/projects/jkudish-jev-mcp): Proof of concept MCP for Typesafe's new Jev AI model
- [pi-warden](https://jevlibrary.dev/projects/devmortimer-pi-warden): Guardrails for Pi built on pi-typesafe that steer the agent instead of interrupting you: Jev judges irreversible and off-task tool calls, detects stuck loops, checks unverified done claims, flags slop
- [TypeSafe MCP](https://jevlibrary.dev/projects/itsmostafa-typesafe-mcp): Bring typed Jev judgments to your coding assistant through a single MCP server.
- [pi-jev (y0usaf)](https://jevlibrary.dev/projects/y0usaf-pi-jev): TypeSafe Jev as a decision layer for the Pi coding agent: a measured tool-call gate plus jev_ask for typed, calibrated answers
- [jev-axi](https://jevlibrary.dev/projects/shiftynick-jev-axi): Agent-ergonomic CLI for TypeSafe's Jev: fast calibrated judgments (pick, rate, check, rank, triage, guard) from the shell
- [jev-codex-router](https://jevlibrary.dev/projects/0xnatoshi-jev-codex-router): Per-turn model & reasoning routing for Codex, driven by Jev (TypeSafe System One): picks the model, thinking depth and speed mode for every turn.
- [jev-mcp (blakestone-x)](https://jevlibrary.dev/projects/blakestone-x-jev-mcp): MCP server for TypeSafe Jev: typed classify, score, check, match and screen for any agent, with confidence on every answer
- [typesafe-cli (geilt)](https://jevlibrary.dev/projects/geilt-typesafe-cli): CLI and agent skill for TypeSafe System One (Jev): typed Choice, Score, and Noul judgments.
- [All 168 entries in Agent tooling](https://jevlibrary.dev/categories/agents)

## Browser & computer use (71 entries)

Browser, desktop, and mobile automation with Jev choosing the action.

- [Jev Ultrafast](https://jevlibrary.dev/projects/browser-use-jev-ultrafast): A browser agent that turns a single goal into a sequence of lightning-fast actions.
- [typesafe-computer-use](https://jevlibrary.dev/projects/awlevin-typesafe-computer-use): Computer use for about $0.0002 a step: OCR the screen, classify the next action with TypeSafe, click. macOS.
- [jev-browser (vlad-terin)](https://jevlibrary.dev/projects/vlad-terin-jev-browser): Jev-powered element selection for your agent’s existing computer-use tools
- [unclutter](https://jevlibrary.dev/projects/kitze-unclutter): WXT browser extension: Jev-powered page clutter removal with reusable template rules.
- [Mobile Jev](https://jevlibrary.dev/projects/droidrun-mobile-jev): An Android agent powered by Jev decisions, with a live studio to watch every step.
- [jev-ego](https://jevlibrary.dev/projects/romaluev-jev-ego): TypeScript browser agent for ego lite: Jev Ultrafast indexed actions, TypeSafe Jev decisions, persistent observe/act CLI. No Chrome or Playwright.
- [psearch](https://jevlibrary.dev/projects/komikat-psearch): Parallel web search for terminals and agents, with local Chromium and Jev-guided exploration.
- [jev-browser (Ying-Kai-Liao)](https://jevlibrary.dev/projects/ying-kai-liao-jev-browser): Browser automation where an LLM plans and Jev (Typesafe System One) decides. Library, CLI and MCP server.
- [computer_use](https://jevlibrary.dev/projects/paulsmith-computer-use-jev): macOS computer use driven by Jev (TypeSafe System One) as the decision maker
- [sift](https://jevlibrary.dev/projects/tylergibbs1-sift): Chrome extension that re-ranks Google results with TypeSafe Jev and folds away sales pages and SEO filler.
- [almond-fastloop](https://jevlibrary.dev/projects/eriestra-almond-fastloop): Almond-fastloop: Almond's browser computer-use rig (Chrome DevTools + TypeSafe Jev), and the Browser Use Olympics benchmark it is measured on.
- [jev-mobile](https://jevlibrary.dev/projects/friedjof-jev-mobile): Fast structured Android control loops with TypeSafe Jev and Mobile MCP
- [All 71 entries in Browser & computer use](https://jevlibrary.dev/categories/browser)

## Applications (90 entries)

Products, tools, and pipelines that call Jev.

- [Jev Trader](https://jevlibrary.dev/projects/jarrodwatts-jev-trader): One trading decision every Monad block. An experiment in real-time market intelligence.
- [notra](https://jevlibrary.dev/projects/usenotra-notra): Marketing analytics platform whose feature flag routes brand-visibility classifiers off an LLM and onto Jev boolean decisions.
- [commit-miner](https://jevlibrary.dev/projects/devanshbatham-commit-miner): Classify Git commit diffs and messages with Jev. Bug fixes, security fixes/CWEs, and change types.
- [blink](https://jevlibrary.dev/projects/ellipsis-dev-blink): Codebase search powered by Jev from @typesafe-ai
- [Jev-Moderation-Bot](https://jevlibrary.dev/projects/brainstormity-jev-moderation-bot): Real-time Discord moderation bot: Jev evaluates messages and metadata in parallel to catch phishing, spam, and social engineering with a progressive escalation ladder.
- [ai-elo-ranker](https://jevlibrary.dev/projects/opaielsheikh-ai-elo-ranker): High-speed recursive AI Elo tournament engine powered by Jev and Swiss matchmaking
- [typesafe-cli (y0usaf)](https://jevlibrary.dev/projects/y0usaf-typesafe-cli): Ask Jev typed questions from the shell: noul, choice, and score answers as numbers, not prose
- [jevmeter](https://jevlibrary.dev/projects/chetaslua-jevmeter): Put a live Jev (TypeSafe) meter on any video: every sentence scored, rendered as a 16:9 edit
- [jevlogs](https://jevlibrary.dev/projects/reachjalil-jevlogs): Open-source Jev log triage for OpenTelemetry. Score the signal before expensive LLM analysis.
- [semdecide](https://jevlibrary.dev/projects/sharziki-semdecide): Typed semantic decisions for Unix pipelines and CI, powered by TypeSafe AI Jev.
- [jevibe-check](https://jevlibrary.dev/projects/sriganesh-jevibe-check): A live tone labeler for Bluesky posts and drafts, using TypeSafe's Jev API.
- [commentcop](https://jevlibrary.dev/projects/ntedvs-commentcop): Put your code comments on trial. Powered by Jev.
- [All 90 entries in Applications](https://jevlibrary.dev/categories/apps)

## Games & simulations (61 entries)

Games and simulations with Jev making the moves.

- [Jev plays Mario](https://jevlibrary.dev/projects/fhshaik-typesafe-mario): Less talking. More jumping. Watch Jev navigate Super Mario Bros. from game state.
- [jev-drone](https://jevlibrary.dev/projects/romanslack-jev-drone): Camera-only autonomous drone in MuJoCo with a small judgment model (TypeSafe Jev) in the loop at 2.5Hz
- [typesafe-snake](https://jevlibrary.dev/projects/sorrycc-typesafe-snake): Snake auto-played by TypeSafe's Jev model: one System One choice per tick, legal moves and facts generated in code
- [tsai-sc](https://jevlibrary.dev/projects/phyous-tsai-sc): TypeSafe Jev controls original StarCraft shareware through keyboard and mouse with recorded action probabilities.
- [mario-jev](https://jevlibrary.dev/projects/shantanugoel-mario-jev): Python prototype that plays NES Super Mario Bros. from structured RAM observations, with Jev answering focused movement and jump questions.
- [rubikjev](https://jevlibrary.dev/projects/0xtrou-rubikjev): Challenge the Jev's intelligence in Rubik Cube puzzles
- [OneVOneJev](https://jevlibrary.dev/projects/emrickgarrett-onevonejev): 1v1 Jev quickscope arena — Three.js + TypeSafe System One
- [casse-brique-typesafe](https://jevlibrary.dev/projects/para-fr-casse-brique-typesafe): A Next.js brick breaker whose paddle is controlled in real time by TypeSafe AI's Jev model. Built with Claude Code.
- [pong-jev](https://jevlibrary.dev/projects/safzanpirani-pong-jev): TypeSafe's Jev plays Atari Pong. One typed Choice question per frame, no coordinates sent to the model.
- [cyber-breach-jev](https://jevlibrary.dev/projects/rchovatiya88-cyber-breach-jev): Cyber-Breach: The Jev Protocol - A tactical cyberpunk arena combat game powered by TypeSafe AI Jev System One decision model
- [Agent-JEV-Tetris](https://jevlibrary.dev/projects/yasserbhb-agent-jev-tetris): using the new model JEV to play the game tetris
- [last-exit](https://jevlibrary.dev/projects/0x963d-last-exit): A cyberpunk border encounter powered by TypeSafe Jev. Bluff the guard. Inspect the receipts.
- [All 61 entries in Games & simulations](https://jevlibrary.dev/categories/games)

## Demos & playgrounds (73 entries)

Live demos and playgrounds.

- [jev-me](https://jevlibrary.dev/projects/jon-devlapaz-jev-me): A grill-me style interrogation of your idea, with Jev doing the grilling.
- [typesafe-ai-playground (BunsDev)](https://jevlibrary.dev/projects/typesafeai-typesafe-playground): Community TypeSafe AI playground: 110 use cases, games, dilemmas and model challenges, with editable prompts, A/B comparisons and a mobile-friendly UI.
- [jev-system-one](https://jevlibrary.dev/projects/haseeb-heaven-jev-system-one): A polished OpenAI + TypeSafe Jev terminal interface for answers with transparent decision reports
- [jev-playground (Little-Planet-Labs)](https://jevlibrary.dev/projects/little-planet-labs-jev-playground): A small Next.js app for experimenting with TypeSafe AI's Jev model (System One)
- [clarity-judge](https://jevlibrary.dev/projects/typesafeai-clarity-judge): Multi-axis writing quality checker powered by TypeSafe AI's Jev model. Separate named checks, each with its own verdict and confidence.
- [gpt-vs-jev](https://jevlibrary.dev/projects/tanaypadar-gpt-vs-jev): Compare GPT generated language with JEV structured Noul decisions on the same input.
- [harden-jev-decides](https://jevlibrary.dev/projects/tylerjharden-harden-jev-decides): JEV picks which stream idea becomes the live MVP. TypeSafe System One decision board.
- [typesafe-ai-playground (markjaquith)](https://jevlibrary.dev/projects/markjaquith-typesafe-ai-playground): A playground for experiments around Jev, TypeSafe's System One model.
- [got-jev](https://jevlibrary.dev/projects/phureewat29-jev-got): Jev (TypeSafe AI) PoC through Game of Thrones
- [jev-should-i-apply](https://jevlibrary.dev/projects/cardotrejos-jev-should-i-apply): Typesafe/Jev public X demo
- [jev-user-jury](https://jevlibrary.dev/projects/cardotrejos-jev-user-jury): Typesafe/Jev public X demo
- [jev-ad-preflight](https://jevlibrary.dev/projects/cardotrejos-jev-ad-preflight): Typesafe/Jev public X demo
- [All 73 entries in Demos & playgrounds](https://jevlibrary.dev/categories/demos)

## Benchmarks & research (155 entries)

Benchmarks, evals, calibration studies, and open replicas.

- [SemIf](https://jevlibrary.dev/projects/theoleecj-semif): Semantic ifs from open models, on a 3090 at home. Independent; not affiliated with Jev or TypeSafe.
- [jev-eval-agent](https://jevlibrary.dev/projects/vinilana-jev-eval-agent): Personal-assistant agent built on Vercel's eve with 100 mocked tools, measuring how many steps it takes when Jev picks the tool versus the LLM.
- [typesafe-ai-benchmark](https://jevlibrary.dev/projects/iammrduncan-typesafe-ai-benchmark): This is a LLM Gateway that mimics typesafe ai structured output. Like an imposter Jev.
- [jev-on-a-laptop](https://jevlibrary.dev/projects/rorshopping-jev-on-a-laptop): Unofficial study: Jev-style parallel typed decisions on stock 1.5B-8B models on an Apple Silicon laptop. Benchmarks, research notes, and a Hugging Face Space demo.
- [open-jev (JoshuaSP)](https://jevlibrary.dev/projects/joshuasp-open-jev): Typed JSON inference with DiffusionGemma, with Every and Jev benchmark results
- [openvons](https://jevlibrary.dev/projects/genai-craft-openvons): openvons (open-Jev): 有限選択肢に確率で答える判断層 — テキスト / 画像 / 日本語音声コマンド
- [jev-benchmarks](https://jevlibrary.dev/projects/abdelstark-jev-benchmarks): Probability-aware evaluation for typed decision models: calibration, selective risk, latency, and reproducible benchmarks.
- [open-jev (daseinlabs)](https://jevlibrary.dev/projects/daseinlabs-open-jev): One-pass option scoring with a local Gemma 3 4B on Apple silicon via MLX, inspired by jevlike, with a Doom demo.
- [jev-lm](https://jevlibrary.dev/projects/y0usaf-jev-lm): A word-level language model whose output layer is Jev: n-gram drafter, Noul chunk verification, bits-per-token eval
- [decider](https://jevlibrary.dev/projects/mapika-decider): One-pass typed decisions with calibrated probabilities (System One style model), fine-tuned from Qwen3.5-2B
- [LegalForecastBench](https://jevlibrary.dev/projects/johnhughes3-legalforecastbench): LegalForecast-MTD benchmark alpha and official evaluation workflows
- [jev-korean-benchmark](https://jevlibrary.dev/projects/mahlernim-jev-korean-benchmark): Reproducible early-access evaluation of Jev on Korean understanding and medical text, with runtime and cost evidence
- [All 155 entries in Benchmarks & research](https://jevlibrary.dev/categories/research)

## Other lists (27 entries)

Other curated lists.

- [awesome-typesafe](https://jevlibrary.dev/resources/abdelstark-awesome-typesafe): Curated list of official resources and community projects for TypeSafe, System One models, and Jev, with a GitHub Pages site.
- [awesome-jev (AnotiaWang)](https://jevlibrary.dev/resources/anotiawang-awesome-jev): A curated list of awesome Jev / TypeSafe System One applications, libraries, and resources.
- [awesome-jev (yibie)](https://jevlibrary.dev/resources/yibie-awesome-jev): A curated list of public projects, integrations, and discussions built on Jev — TypeSafe AI's System One model for typed decisions.
- [awesome-jev](https://jevlibrary.dev/resources/cobanov-awesome-jev): A curated, source-backed list of projects built with Jev, TypeSafe AI's System One model for typed decisions.
- [awesome-jev (fatwang2)](https://jevlibrary.dev/resources/fatwang2-awesome-jev): A source-backed Jev project directory with a reusable Jev-only GitHub review workflow.
- [awesome-jev-projects](https://jevlibrary.dev/resources/logicrw-awesome-jev-projects): Awesome Jev: source-backed open-source ecosystem radar, plain-language project discovery, and automatic GitHub sync
- [awesome-jev (OmniJev)](https://jevlibrary.dev/resources/omnijev-awesome-jev-gallery): Papers, open reproductions and independent evaluations behind System One models and Jev.
- [awesome-jev-typesafe](https://jevlibrary.dev/resources/valentynkit-awesome-jev-typesafe): Typed decisions with TypeSafe's Jev, the first System One model
- [awesome-jev (kraayenjon)](https://jevlibrary.dev/resources/kraayenjon-awesome-jev): A curated list of Jev use cases, projects, SDKs, and resources. Jev is TypeSafe AI's System One model for fast, typed decisions in software — Choice, Score, and Noul with calibrated probabilities.
- [awesome-jev-zh](https://jevlibrary.dev/resources/yzfly-awesome-jev-zh): Jev / TypeSafe System One 中文精选列表：官方资料、SDK、爆款应用、Agent 工具、开源复现与独立评测，附中文上手指南，每日自动收录 GitHub 热门项目。
- [awesome-jev-tools](https://jevlibrary.dev/resources/v-modal-awesome-jev-tools): A curated list of tools built for Jev — TypeSafe AI's System One model for typed decisions.
- [awesome-jev (AppitStudio)](https://jevlibrary.dev/resources/appitstudio-awesome-jev): Curated Jev resources and runnable examples for typed AI decisions.
- [All 27 entries in Other lists](https://jevlibrary.dev/categories/lists)

## Articles & threads (50 entries)

Coverage, write-ups, and X threads.

- [Hacker News launch thread](https://jevlibrary.dev/resources/hacker-news-launch-thread): 1,800-point thread debating whether typed decisions replace LLM calls for classification, routing, and scoring.
- [AINews: Jev, a System One Model that only decides](https://jevlibrary.dev/resources/ainews-jev-a-system-one-model-that-only-decides): Latent Space's launch-day roundup: over 100x faster and 200x cheaper than small frontier LLMs.
- [TypeSafe AI debuts model for machines that plays Doom](https://jevlibrary.dev/resources/typesafe-ai-debuts-model-for-machines-that-plays-doom): The Register on the launch, the Doom demo, and the $40M seed round.
- [Jev: System One models explained](https://jevlibrary.dev/resources/jev-system-one-models-explained): The Neuron's explainer on AI decisions without a chatbot.
- [Mini-Vibe Check: Jev judged everything I have written in 0.7 seconds](https://jevlibrary.dev/resources/mini-vibe-check-jev-judged-everything-i-have-written-in-0-7-seconds): Every's Mike Taylor runs his whole archive through Jev.
- [Jev: The Language Model That Will Not Talk](https://jevlibrary.dev/resources/jev-the-language-model-that-will-not-talk): Anthony Maio's essay on what a model that cannot generate text is for.
- [TypeSafe Jev: the first decision-only model class](https://jevlibrary.dev/resources/typesafe-jev-the-first-decision-only-model-class): Release-week technical roundup: API, evals, adapter, and skill.
- [He says he co-invented ChatGPT. His new AI will not write a word](https://jevlibrary.dev/resources/he-says-he-co-invented-chatgpt-his-new-ai-will-not-write-a-word): dev.to walkthrough of the Vercel AI SDK evaluate integration.
- [Typed decisions, not chat](https://jevlibrary.dev/resources/typed-decisions-not-chat): Independent walkthrough separating TypeSafe's published claims from public evidence.
- [What is Jev?](https://jevlibrary.dev/resources/what-is-jev): Short practical intro with a Python ticket-triage example.
- [AI that does not talk](https://jevlibrary.dev/resources/ai-that-does-not-talk): Practical guide: playground, Python and JS SDKs, raw HTTP, and the agent skill.
- [Jev vs Mistral and Gemini for event validation](https://jevlibrary.dev/resources/jev-vs-mistral-and-gemini-for-event-validation): Head-to-head test at validating local event listings, with cost and latency.
- [All 50 entries in Articles & threads](https://jevlibrary.dev/categories/articles)

## Jev for marketing (6 entries)

Content, ads, brand visibility, and growth experiments.

- [notra](https://jevlibrary.dev/projects/usenotra-notra): Marketing analytics platform whose feature flag routes brand-visibility classifiers off an LLM and onto Jev boolean decisions.
- [draftpulse](https://jevlibrary.dev/projects/pekth-draftpulse): Experimental: live X draft viral scorer powered by TypeSafe Jev
- [jev-seo](https://jevlibrary.dev/projects/akashpriyadarshii-jev-seo): 100% free ₹0 agent-first SEO & GEO CLI suite and MCP server in Rust replacing Semrush and OpenSEO via DuckDuckGo and TypeSafe Jev System One
- [SuperX · post scoring](https://jevlibrary.dev/projects/superx-post-scoring): A reported Jev workflow that evaluates draft X posts across multiple questions, helping creators compare and refine their writing before publishing.
- [StealAds · ad intelligence](https://jevlibrary.dev/projects/stealads-ad-intelligence): A demonstrated Jev workflow for breaking down competitor ads by hook, format, offer, CTA, and landing-page fit. The creator announced plans to bring the workflow to StealAds and MCP; current availability has not been verified.
- [Finding what makes content work](https://jevlibrary.dev/projects/finding-what-makes-content-work): An author-reported experiment using Jev to classify past X posts by topic, hook, and tone, then compare those attributes with engagement. A content-research experiment, not a standalone app or a guarantee of growth.