Launches on September 19, 2026

A single-page Next.js app that classifies PolyAI BANKING77 customer support queries with Jev, TypeSafe AI's System One model, and streams the results into the UI one ticket at a time.

BrowseComp-Plus with a cheap LLM and Jev (TypeSafe's non-generative classifier): agent pipeline, the jevlog language, and every experiment

Game intuition service that uses typesafe-ai/jev instead of a behavior tree.

Zero-Shot Classification of Protein Binders with Jev

Jev-style typed decisions (semantic ifs) from a local Ternary Bonsai 2 27B on Apple Silicon: one forward pass, no generation. Independent; not affiliated with TypeSafe/Jev.

Discord bot that scores how likely a statement is to be true, using TypeSafe's jev model

TypeSafe Jev predicts a synthetic 32-team tournament in round fan-out: calibrated Brier vs Elo/seed/oracle plus Monte Carlo champion odds.

ego-browser + TypeSafe Jev: per-step browser decisions at ~1s and near-zero cost. Skill, router pilot, and measured results.

Jev based exam paper categorisation used on matura.lol

A traffic city where every car is driven by TypeSafe's Jev model, benchmarked against a rule-based driver.

Benchmarking TypeSafe Jev against 15 chat-model configurations on support-ticket classification: latency, tokens, cost, accuracy.

Classify companies and job titles with Jev (TypeSafe AI) — calibrated probabilities at $0.042 per million input tokens, output free

Claude Code skill for TypeSafe's Jev decision model — typed decisions via TypeSafe, Vercel AI Gateway or OpenRouter

On-demand Jev (TypeSafe System One) question catalog for CMS-style content — classify, gate, and judge your own content by id.

Logic gates for AI judgment.

Backs the WordPress AI plugin's Comment Moderation with TypeSafe Jev, through Connector for TypeSafe Jev

Corpus curation with TypeSafe Jev — pass/fail gates, curated vs rejected JSONL

LLM-based permission gate for OpenCode agents: Jev auto-approves routine tool calls and fails open to a human prompt on anything risky or uncertain.

A provider-agnostic AI decision gateway using TypeSafe AI's Jev (System One) model to gate, route, and verify expensive LLM workflows

Decision Maker at Meteora - a decision agent for Meteora DLMM liquidity on Solana. Named after Jev, TypeSafe AI's System One model (https://console.typesafe.ai/home), which answers its typed Noul and Score questions. Ranks the live pool universe, runs live rug cards, and holds 3-5 pools under explicit hard gates and book rules.

Demo for new jev

Find underpriced GoDaddy expiry domains with deterministic market signals and TypeSafe JEV judgments.

A 3D driving simulator where Jev, TypeSafe's decision model, chooses what the car does. Code eye or Gemini camera eye, code reflexes, an evaluation harness.

Email triage with Jev (TypeSafe) over the Vercel AI Gateway

Architecture for running many semantic decisions through one validated path, with a labelled 111-case benchmark comparing TypeSafe Jev against a Claude baseline, and a dashboard for inspecting any single decision. Experimental, not production.

Classification benchmark: Jev vs Claude Sonnet 5 vs Claude Opus 5 on AG News

Benchmarking TypeSafe Jev against general-purpose LLMs on support-ticket routing, with a focus on latency, accuracy, and confidence.

Small experiments with Jev by TypeSafe

TypeSafe Jev plays Guideline Tetris through Vercel AI Gateway.

Calibrated PR review gates powered by TypeSafe Jev: a GitHub Action, a local CLI, and an OpenCode plugin

Reproducible TypeSafe Jev vs Knowledgator GLiClass bake-off on LocalLLaMA/typed-decisions

Standalone Go CLI and MCP server for TypeSafe Jev: typed judgments, JSONL evaluation, and resumable batches.

Measuring TypeSafe AI's Jev on 41-clause contract review (CUAD, 20,500 decisions) against fast, cheap LLMs — latency, cost and F1

Typed decision control plane for agents, powered by TypeSafe Jev

TypeSafe Jev (System One) integrations for OMP, MCP, Claude Code, and Pi — fail-open decision routing, gating, and verbatim compaction.

Research pipeline testing whether TypeSafe's Jev (via Vercel AI Gateway) can judge news and market data fast enough to matter. Bitcoin and US stocks, paper trading only.

HS code assistant MVP — localhost demo for customs HS6 suggestions

Run iOS Simulator goals with Jev, compare decision models, and replay every attempt. Python CLI and Brigade host wrapper.

DOM-native browser agent tested live against IRCTC train search

Evaluate Jev (TypeSafe AI) on Japanese customer-inquiry data: 150-item dataset + comparison script (Jev / LLM / rule-based)

Experiments with TypeSafe's Jev: triage benchmark vs LLM, and a source study of jev-ultrafast vs Cline's jev-browser

Small apps for evaluating TypeSafe AI's System One model (Jev). Unofficial.

Sandbox for TypeSafe Jev: typed judgments over inbound messages

Triage de leads de WhatsApp/CRM con Jev (TypeSafe AI): ruteo por confianza, plantilla n8n y benchmark en español. Sin dependencias. No afiliado.

Do I need to look at what Claude Code just did? A calibrated verdict per stop, judged by TypeSafe Jev. Pairs with jev-lens.nvim

Voice loan-call triage using TypeSafe Jev: intent, info sufficiency and a lending decision from typed AI judgments

An LSP server whose ambient pass runs the rules a repository states in `.jev/rules/*.json`: each candidate line is sent to Jev, a classifier that answers one typed question with a value and a probability, never prose.

Measured cost, latency and raw output from the live Jev API (TypeSafe AI System One model) across 8 use cases — reproducible

Classifies PRs as "merge candidate" or "human review" required using Jev's typed decision making.

Hands-on starter kit for TypeSafe Jev: what it can do, what to install, runnable examples, a jev CLI with a viewer, and a bilingual tutorial