A playful collection of decision demos spanning dispatch, moderation, search, desktop control and a 32-agent arena.
ACTUAL RECORDING12 SEC ↗
The directory02 / 02 · Experiments
Find something worth exploring.
Every category, from games and tools to repositories and writeups, traced to a primary source.
Same ideas.
Different paths.
This is the full registry, not just the featured picks. Figures like speed, cost, and stars are maker-reported or captured snapshots, not JevMade measurements.
1610 experiments · showing 1–60
Browser Use's agent asks Jev to choose the next browser operation and target element from a fresh table on every step.
SOURCE SCREENSHOTEXPAND ↗
from the source
git clone https://github.com/browser-use/jev-ultrafast.gitcd jev-ultrafastuv synccp .env.example .env# Add TYPESAFE_API_KEY and TEXT_MODEL_API_KEY.
Speak an English or Ukrainian drawing command while pointing, and Jev translates it into an action on a tldraw canvas.
ACTUAL RECORDING12 SEC ↗
Jev plays headless Doom from object bearings, distances, health, and ammo rather than game frames.
ACTUAL RECORDING12 SEC ↗
Jev plays NES Super Mario Bros. by choosing controls from emulator state rather than pixels.
SOURCE SCREENSHOTEXPAND ↗
from the source
NES emulator -> telemetry/RAM parser -> structured JSON -> Jev Choice -> controller inpu...
A simulated Panda robot asks Jev how to move a block over a barrier.
ACTUAL RECORDING12 SEC ↗
Jev picks a flight path through each pipe, then Flappy Bird turns that route into precisely timed flaps.
ACTUAL RECORDING12 SEC ↗
Jev steers a Pac-Man recreation using structured game state.
ACTUAL RECORDING12 SEC ↗
Jev places Tetris pieces by considering holes, stack height and the current board.
ACTUAL RECORDING12 SEC ↗
World Monitor benchmarks Jev's headline severity and topic labels against cached labels and a saved judged set.
SOURCE SCREENSHOTEXPAND ↗
from the source
async function callJev(titles) { const t0 = performance.now(); for (let attempt = 0; ; attempt++) { const r = await fetch(JEV_ENDPOINT, { method: 'POST',
DeerFlow includes opt-in Jev extensions for text classification, prompt-injection screening, and pruning old tool results.
SOURCE SCREENSHOTEXPAND ↗
from the source
async def _probability(self, client: httpx.AsyncClient, key: str, excerpt: str) -> f... body = { "model": self.options.model, "state": {"content": excerpt}, "questions": {"injection": _question()},
Oh My ClaudeCode can compare Jev with its existing coding-agent heuristics in shadow mode, then optionally let Jev make those decisions while retaining the original behavior as a fallback.
SOURCE SCREENSHOTEXPAND ↗
from the source
export async function queryJev( state: unknown, questions: JevQuestions, options: JevClientOptions,): Promise<JevResponse> {
Oh My Pi lets coding agents use Jev for structured decisions such as choosing an option or scoring a candidate.
SOURCE SCREENSHOTEXPAND ↗
from the source
async #attempt<T>(path: string, body: string, key: string, signal: AbortSignal | undefi... const url = `${this.baseUrl}${path}`; const headers: Record<string, string> = { ...this.#headers, Authorization: `Bearer ${key}`,
A Composio provider that asks Jev which tool to use and which arguments to supply.
SOURCE SCREENSHOTEXPAND ↗
from the source
decision = provider.decide(tool_set, REQUEST) if decision["kind"] == "abstain": print(f"Jev abstained: {decision['reason']}") if decision["risk"] != "read_only": print(f"Not executing {decision['tool']}: its risk class is {decision['risk']}")
Demonstrates a bounded Jev action chooser inside Cua Driver loops, with mock and credentialed live paths.
SOURCE SCREENSHOTEXPAND ↗
Run Convai Innovations' open-weight typed-decision model locally with a Jev-shaped request format; Laya is an independent model, not TypeSafe's Jev.
SOURCE SCREENSHOTEXPAND ↗
jcode lets its fast browser agent ask Jev to choose the next move from the actions currently on screen.
SOURCE SCREENSHOTEXPAND ↗
from the source
async fn decide(&self, request: &DecisionRequest) -> Result<Decision> { let body = request_body(request)?; let questions = body["questions"] .as_object() .context("Browser decision questions are missing")?
QuantDinger can ask Jev to review a live trade entry before the order proceeds.
SOURCE SCREENSHOTEXPAND ↗
from the source
response = requests.post( url, headers={"Authorization": f"Bearer {config['api_key']}"}, json={"model": model, "state": self._state_payload(request), "questions": JEV_QUESTI...)
OpenChamber asks Jev to choose a model for Auto sessions and, when enabled, flag risky tool permissions for human review.
SOURCE SCREENSHOTEXPAND ↗
from the source
export const createJevClient = ({ fetchImpl = fetch, timeoutMs = JEV_TIMEOUT_MS } = {})... /** Resolves to the parsed answers; throws with `status` on an HTTP error and `code: '... ask: async (request, token) => { const abort = new AbortController(); const timer = setTimeout(() => abort.abort(), timeoutMs);
Sure can ask Jev to sort bank transactions into a family's categories and benchmark those choices against labeled examples.
SOURCE SCREENSHOTEXPAND ↗
from the source
def decide!(state:, questions:, model: "") raise Error, "No questions provided" if questions.blank? questions = questions.transform_keys(&:to_s) questions.each { |key, question| validate_question!(key, question) }
GenOffice can ask Jev to reorder local search results by how well each document excerpt matches a query.
SOURCE SCREENSHOTEXPAND ↗
from the source
export const fetchTransport: JevTransport = async (url, body, key, signal) => { const res = await fetch(url, { method: 'POST', headers: { Authorization: `Bearer ${key}`, 'Content-Type': 'application/json' }, body,
Train and self-host Qwen-based decision models that answer Jev-style typed questions in one pass; kev is an independent implementation, not Jev.
SOURCE SCREENSHOTEXPAND ↗
Reads supported Android chat screens, judges conversational intent, and offers replies that users may fill before sending manually.
SOURCE SCREENSHOTEXPAND ↗
Vercel's open framework for building agents.
SOURCE SCREENSHOTEXPAND ↗
An agentgateway example that screens LLM requests and responses with Jev guardrails.
SOURCE SCREENSHOTEXPAND ↗
Qlty Slop One asks Jev to score code-quality traits in source excerpts, then rolls those scores into per-file reports.
SOURCE SCREENSHOTEXPAND ↗
from the source
/// The validated answers for one excerpt, from the cache when the exact /// request was answered before. pub fn request(&self, state: &RequestState) -> Result<Answers> { let body = state.body()?; let key = canonical::digest(&body)?;
A Claude Code plugin that uses Jev to decide which context survives compaction.
SOURCE SCREENSHOTEXPAND ↗
from the source
claude plugin marketplace add tamaratran/fast-jev-compactionclaude plugin install fast-jev-compaction@fast-jev-compaction
NanoJev trains a 0.6B parallel decision model and compares it with Jev across maze, Snake, and ViZDoom tasks.
SOURCE SCREENSHOTEXPAND ↗
An AI stock-research team that can use Jev to rate an investment and flag risks before writing its report.
SOURCE SCREENSHOTEXPAND ↗
from the source
"rating": { "type": "choice", "instructions": "综合团队讨论结论，该股票的最终投资评级是哪一档？", "criteria": dict(RATING_MAP), },
Nimble compares Jev's typed probabilities with locally trained schema adapters on curated decision datasets.
SOURCE SCREENSHOTEXPAND ↗
from the source
payload = request_payload(row, model)response = request_with_retry(transport, payload, attempts=attempts, sleep=sleep)validate_teacher(row, response, model)return jev_row(row, response, time.perf_counter() - started)
LLM Gateway uses Jev to route model requests and check text against moderation categories.
SOURCE SCREENSHOTEXPAND ↗
from the source
const score = answers[category]?.noul;const matched = score > threshold;categories[category] = matched;
SemIf runs open models locally as probabilistic semantic conditionals on an RTX 3090.
SOURCE SCREENSHOTEXPAND ↗
Play Snake or request typed decisions from Core ML ports of Laya on Apple hardware; the models are open-weight alternatives, not Jev.
SOURCE SCREENSHOTEXPAND ↗
Drive a Mac through its accessibility tree, with Jev choosing an operation and a live element on each turn.
ACTUAL RECORDING12 SEC ↗
from the source
const req = buildRequest("save the file", { app: "TextEdit", window: "Untitled" }, offer...assert.deepEqual(Object.keys(req.questions), ["target", "command", "present", "destructi...assert.equal(Object.keys(req.questions.target.criteria).at(-1), "none");assert.equal(req.questions.present.type, "noul");assert.equal(req.questions.target.criteria[`@${S}:e3`].what, 'button "Save"');
Mediary Scout filters irrelevant and adult-content media before those candidates reach more expensive workflow stages.
SOURCE SCREENSHOTEXPAND ↗
from the source
if (score === undefined) return true;const band = classifyJevScore(score);if (band === "drop") return false;
Symfony's first-party bridge for asking Jev several choice, score and yes-or-no questions about one shared state.
SOURCE SCREENSHOTEXPAND ↗
Kuest asks Jev to spot unclear prediction-market rules and rank useful context for a draft.
SOURCE SCREENSHOTEXPAND ↗
from the source
export async function requestOpenRouterDecisions( request: OpenRouterDecisionRequest, options?: { apiKey?: string; timeoutMs?: number },): Promise<OpenRouterDecisionResponse> { const apiKey = options?.apiKey
Jevlike trains a compact model to choose among a list of text options that can change from request to request.
SOURCE SCREENSHOTEXPAND ↗
from the source
uv venvsource .venv/bin/activateuv pip install -e '.[dev]'jevlike-data synthetic --output data/syntheticjevlike-train data/synthetic/train.jsonl \
A trading bot asks Jev what to do on every Monad block.
SOURCE SCREENSHOTEXPAND ↗
A Vercel Labs command-line tool for generating content from the terminal.
SOURCE SCREENSHOTEXPAND ↗
choice
Which team should handle this request?
Run a local, non-autoregressive System One model behind Python and JavaScript clients compatible with Jev’s request shape.
SOURCE SCREENSHOTEXPAND ↗
A macOS menu-bar companion that turns a typed click task into a short sequence over locally detected controls.
SOURCE SCREENSHOTEXPAND ↗
from the source
"absent": .noul( instructions: "Is the control needed to make the next bit of progress on... ), "kind": .choice( instructions: "How should that element be acted on?",
Type one thought and watch the text box morph into an event card, checklist, conversion, calculation, or other fitting tool.
SOURCE SCREENSHOTEXPAND ↗
Pi Typesafe Compact asks Jev which parts of a coding conversation deserve space in the compacted summary.
SOURCE SCREENSHOTEXPAND ↗
from the source
const selectedUnits = evaluation.decisions .filter((decision) => decision.summarize) .map((decision) => decision.unit);
Lynkr uses Jev as a second opinion when choosing a model tier for a coding request, adjusting its own score-based route only when Jev's answer is confident enough.
SOURCE SCREENSHOTEXPAND ↗
from the source
if (typeof jevConfidence !== 'number' || jevConfidence < JEV_PROMOTE_CONFIDENCE) return...if (_pri(jevTier) < _pri(base)) return { tier: jevTier, score: TIER_MIDPOINT[jevTier] ??...
Astra plans a Minecraft route while JEV chooses bounded actions for a recorded, source-verifiable playthrough.
SOURCE SCREENSHOTEXPAND ↗
Uses local OCR and Jev to judge Windows chat messages, rank three drafted replies, and fill one without sending it.
SOURCE SCREENSHOTEXPAND ↗
Simple Jev exposes open Hugging Face models as structured classifiers and includes playgrounds, evaluations, and an autonomous-driving demo.
SOURCE SCREENSHOTEXPAND ↗
A cautious Codex computer-use skill that chooses UI targets from accessibility text while local policy guards consequential actions.
SOURCE SCREENSHOTEXPAND ↗
Adds typed probability readouts and calibration heads to ordinary language models, with Jev-style evaluations and saved results.
SOURCE SCREENSHOTEXPAND ↗
Indexes Jev SDKs, integrations, applications, evaluations, and learning material with source-backed annotations.
SOURCE SCREENSHOTEXPAND ↗
Mini Taiwan Pulse tests whether Jev can pick the map layers relevant to a visitor's question.
SOURCE SCREENSHOTEXPAND ↗
from the source
const probabilities = parseNoulAnswers(body, batch.questionIds);for (const candidate of batch.candidates) decisions.set(candidate.key, knownDecision(probabilities.get(questionIdForLayer(candid...
DSCode asks Jev whether a coding agent's risky action should proceed, stop or receive a fuller review.
SOURCE SCREENSHOTEXPAND ↗
from the source
verdict = await jev.approval({ action, context, sessionId: req.agent.session.id, signal:...if (verdict !== undefined) { if (verdict.decision !== 'defer') return applyVerdict({ decision: verdict.decision, re...}
Collects Jev projects, reusable agent skills, and editable decision recipes with an optional validated Python client.
SOURCE SCREENSHOTEXPAND ↗
Coder One uses Jev to steer and check bounded autonomous coding sessions.
SOURCE SCREENSHOTEXPAND ↗
from the source
let mut judge = policy .judge(jev_client, workdir.clone(), &state.issue, recorder.clone()) .with_deep(policy.deep_jev());judge.survey(&mut state).await;
Rizzo Flow serves local Spark models through a Jev-compatible API and visual playground without generating answer text.
SOURCE SCREENSHOTEXPAND ↗
Compare Jev's market probabilities and trading choices in a Polymarket benchmark that never places real-money orders.
SOURCE SCREENSHOTEXPAND ↗
from the source
probability = noul_probability(answers, "probability")action = choice_value(answers, "action")confidence = choice_value(answers, "confidence")
See where a Claude Code session made progress and where it went in circles, with Jev reviewing a redacted trace.
SOURCE SCREENSHOTEXPAND ↗
from the source
key: "progressingEfficiently", label: "Progress", question: "Did the agent make steady, meaningful progress toward the user's task... higher_probability_is_better: true, answer: MetricAnswer::Noul,
Shows intent, risk, and ranked reply suggestions beside supported macOS chats while keeping message sending under user control.
SOURCE SCREENSHOTEXPAND ↗
Browse a bilingual, source-linked index of Jev projects grouped by decision pattern, with dated link and call-site checks.
SOURCE SCREENSHOTEXPAND ↗
Jev, briefly
A decision model, not a chatbot.
A chat model generates text token by token and hopes you parse it. Jev never generates a word. Your code sends the state of the world plus typed questions; Jev returns every answer in one parallel pass — typed values with calibrated probabilities and a confidence score your code can trust. About 70–500 ms end to end, $0.042 per million input tokens, output free. Trained with what TypeSafe calls Reinforcement Learning for Calibrated Decisions.
choice Pick one of these.
You define the options. Jev returns the chosen option, a probability for every option, and a confidence. The workhorse of routing, agents, and games — the answer space is always legal, so there is nothing to parse and nothing to hallucinate.
score Rate this on a rubric.
You describe ordered levels — say trivial / normal / critical. Jev returns the level, the probability of each, and a confidence. Scores turn fuzzy judgment (“how severe is this log line?”) into a number-free decision your code can branch on.
noul Is this statement true?
You assert a statement; Jev returns the probability that it is true — a noul. Moderation, verification, guardrails, “does this diff actually fix the bug?”: one question, one calibrated probability.
Official material lives at typesafe.ai and docs.typesafe.ai. JevMade is an independent community registry — not affiliated with or endorsed by TypeSafe AI.
Recurring lessons
Patterns that keep showing up.
One decision per tick
Games, drones, trading bots, and browser agents all converge on the same loop: serialize the state, ask one decisive question, act, repeat. Jev’s latency makes the loop feel instant — the model lives inside the control loop, not outside it.
Confidence as a gate
The answer says what; the confidence says whether to act. The most reliable entries threshold on confidence to route edge cases to a slower model or a human — automation with an honest escape hatch.
Shrink the choice space in code
Makers rarely ask Jev to pick from everything. Local tactics prune 225 gomoku moves to ~40; DOM filters turn a page into an element table; code narrows options, Jev judges within them.
Jev judges, LLMs talk
A recurring split: Jev makes every decision cheaply and instantly, and a small LLM is only invoked when a human-facing string must actually be written. Decision and generation are separate budgets.
Made something with Jev?
Send the link — repo, live demo, post, or video — plus a line on what it does and which primitives it uses. Every entry is verified against its primary source before it ships.
Submit your experiment → hello@JevMade.com