Launches on September 19, 2026

TypeSafe Jev bids in sealed-bid auctions: threshold fan-out reconstructs a calibrated value CDF, scored by regret and truthfulness.

Multi-dimensional writing voice profiler powered by Jev (TypeSafe AI)

音声でブラウザを操作する Chrome 拡張機能(発話の解釈は TypeSafe の Jev)

Control your browser by voice. A Chrome extension that turns speech into typed commands using TypeSafe's Jev model — which selects from page elements rather than generating selectors.

Japanese speech → local STT → Jev typed decisions → deterministic actions.

TypeSafe AI's Jev vs GPT-5.6 Luna vs Claude Sonnet 5 on support-ticket triage (Vercel AI Gateway)

Zero-shot text classification: TypeSafe Jev vs BERT-family zero-shot (NLI, embeddings) on 6 public tasks + PAWS pairs, with label-equivalence curves and an arXiv contamination control

jeval: open-source evaluations for AI outputs and agents, judged by Jev

Auto-marking maths scripts with Jev (TypeSafe System One): 2,054 scripts, 96.6% agreement with human markers

Record/replay regression tests for Jev (TypeSafe System One) question packs — accuracy, calibration and cost gates in CI

MCP server that lets an LLM agent control a real Android phone, using TypeSafe's Jev to pick real, runtime-discovered targets instead of hardcoded automation.

A desktop home for TypeSafe Jev.

Discover bounded web application action spaces with Playwright and TypeSafe Jev.

Public project submissions for jevfast.com. Issues only; website source is maintained separately.

An infinite feed built from your own browser history, ranked in real time by TypeSafe's Jev. No likes, no follows, no accounts.

grep lines by what they mean

Deterministic decision runtime, zero-token caching, and certainty calibrator for TypeSafe AI (Jev).

grep for meaning: find the error in a 10,000-line log, a commit by description, the command for a task. A Unix CLI, no key needed. Powered by Jev from TypeSafe AI.

simple discord bot that passes a message through some jev prompts

Autonomous browser testing that turns discovered bugs into deterministic regression tests. A model may choose what to try; only browser evidence decides pass or fail. CLI + MCP server.

Playing around with Typesafe's Jev

The jev command-line interface and MCP server for the TypeSafe API.

Local System One for typed decisions. Scores noul, choice, and score from option logits — never free JSON. TypeSafe-compatible POST /v1/systemone with layered routing: on-device LFM2.5 first, optional second encoder, cloud Jev only if local fails. Fine-tune on proprietary labels. Not official Jev.

GMOのFX/BTC市場データに対応したローカル市場研究ターミナル。ライブチャート、ペーパートレード、バックテスト、安全監督、TypeSafe Jev連携。安全機構を整えたうえで実売買対応予定。

jevrun: Linux desktop agent driven by TypeSafe Jev — type a task, Jev drives the computer

Nothing to read: a message only when your app is actually broken, with what kind of problem it is. Log and alert triage powered by TypeSafe Jev.

Jev (TypeSafe System One) research: API notes, benchmarks, community experiments, agent-loop patterns

Probabilistic judgment. Deterministic execution.

Jev-native LLM output judge kit (skill + agents + core)

A Go CLI for TypeSafe Jev judgments, with JSON input/output and optional caching

Planned Rust Laya decision runtime with native Apple Metal and Jev-compatible HTTP API

Chrome extension that blurs low-value LinkedIn feed posts using TypeSafe's Jev model, with one-click reveal

Chrome extension that detects AI-generated LinkedIn posts with TypeSafe's Jev model and slams a SLOP stamp on them

A writing editor that re-scores 15 typed rubric dimensions on every typing pause — one Jev System One call per pause, ~$0.000004 each.

Reproducing jev classifier API on top of llama.cpp

Structured routing demo: TypeSafe's Jev makes one schema-constrained decision picking which of five coding specialists handles your request, then that specialist streams the answer on its own OpenCode Go model.

어르신 프로그램 출석 위험 모니터링 대시보드 · TypeSafe AI JEV 연동

Personal macOS inbox triage app powered by Gmail API and TypeSafe Jev

Decision-native enterprise inbox: typed AI signals, deterministic priority policy, confidence-aware routing.

An original platformer and from-scratch PPO framework, with Jev-assisted learning experiments, Colab training, data, curves and gameplay GIFs.

Use TypeSafe's System One models (Jev) inside any agent

Meridian | AI context-intake and contraction risk & compliance micro-app with live TypeSafe AI vs OpenAI evaluation

Discord bot that evaluates message like chess

OpenAI-compatible proxy: MinusPod ad detection via Jev (TypeSafe decision model)

Interviewer-side live signal layer for mock interviews, built on TypeSafe's Jev. Typed judgments, a suppression policy, and an honest usefulness gate.

TypeSafe jev-powered LLM model router: classify → policy → tiered pools → OpenAI-compatible passthrough proxy (streaming, tools, full agent loops)

Why an agent stopped and how risky its prompt is, as a badge on its pane: TypeSafe's Jev reads the screen modisa's rules can't.

moonbit CI gate with jev

Identify a pressing environmental problem affecting your community or the world and develop a creative solution to address it.

Jev-graded tool approvals for oh-my-pi: suppress the approval prompts you did not need, keep the ones that matter.