Small decisions. Interesting possibilities.Submit contentSubmit
Explore what's possible

Experiments Use Cases with Jev AI

Probes and benchmarks built to reveal a Jev capability or limit.

143 entries
X Post · Experiments

JevCity

Jev selects traffic-light phases in a city-grid simulation.

Cadence not verified
Added
X post ↗
Jev Explained Playground preview
Project · Experiments

Jev Explained Playground

The examples make the boundary between classification and action concrete: Jev supplies typed probabilities, while ordinary application code applies thresholds, confidence checks, and human-review branches.

Runs on demand
Added
See how it works →GitHub ↗
★ 34
X Post · Experiments

Jev Bot Survival Mode

A creator-posted trading experiment shows an autonomous multi-agent dashboard and market activity.

Cadence not verified
Added
X post ↗
X Post · Experiments

Local Audio Switchboard

An audio drop routes language detection, transcription, privacy and task extraction through local models.

Runs on demand
Added
X post ↗
X Post · Experiments · Games

Jev and Opus Tetris API Race

Two models play the same Tetris piece sequence while a live dashboard compares outcomes and response times.

Cadence not verified
Added
X post ↗
X Post · Experiments

Receptionist Outcome Forecasting

Jev forecasts booking outcomes from structured receptionist-call traces and answers five post-call questions.

Runs on demand
Added
X post ↗
X Post · Experiments

Jev + Opus Research Agent

A creator demo pairs Opus-built option sets with repeated Jev decisions inside a research loop.

Cadence not verified
Added
X post ↗
X Post · Experiments

Jevolution

A hackathon ecosystem demo lets AI agents hunt, compete and survive to generate behavioral data.

Cadence not verified
Added
X post ↗Website ↗
X Post · Games · Experiments

Jev Doom Multiverse

A Doom agent tries parallel futures from one checkpoint, compares outcomes, and continues from the best branch.

Cadence not verified
Added
X post ↗
X Post · Experiments

JevForm

A form changes its next question dynamically instead of relying on a large fixed if/then tree.

Runs on demand
Added
See how it works →X post ↗
X Post · Games · Experiments

Jev Catan Bots

Several Jev-controlled players attempt to play and negotiate in a forked Catan game.

Cadence not verified
Added
X post ↗
X Post · Experiments

MOSS Jev Rover

A rover simulation changes litter-picking targets when its instruction changes from cans to bottles.

Cadence not verified
Added
X post ↗
X Post · Experiments

723 ads, 30 personas, 22 cents

jev KILLED the focus group. it scrolled 723 ads as 30 buyer personalities 21,690 stop or scroll decisions. 22 cents. (will be avail in @StealAds + mcp)

Runs on demand
Added
X post ↗
X Post · Experiments

Jevassembler

An experimental terminal demo asks Jev to choose the next CPU-style instruction for a task at runtime.

Cadence not verified
Added
X post ↗
OpenRouter’s Ori Eval preview
X Post · Experiments

OpenRouter’s Ori Eval

1/ Jev, a decision model by @typesafeai, sparked a burst of projects and discussion. We tested it using Ori Eval against popular LLMs on OpenRouter at judging. Jev was >5x faster than the next fastest model, and even its slowest request…

Runs on demand
Added
X post ↗
X Post · Experiments

Jev Accounting Prototype

A creator-posted accounting prototype connects the Jev API to a mock interface.

Cadence not verified
Added
X post ↗
X Post · Experiments

deepseek-v4.1-flash-jev

you can make any open source model behave like jev with just a bit of inference engineering. it's shockingly easy. to prove it, we built a new endpoint we're calling deepseek-v4.1-flash-jev. see the demo below. here's how it's done: sglang…

Runs on demand
Added
See how it works →X post ↗
X Post · Experiments

Jev against Fable 5.1, 100 builders

Jev is INSANE. I asked 100 Indie Hackers to build a post scheduler with: Jev vs Fable 5.1. The results are unexpected 🤯 (You can clearly see Jev is faster at doing stuff)

Runs on demand
Added
X post ↗
X Post · Experiments

Jev Calculator

A real-time calculator demo sends changing interface state to Jev and displays the returned decisions as immediate results.

Cadence not verified
Added
X post ↗
X Post · Experiments

A dual-arm robot with Jev in the middle layer

型付き意思決定AIモデル「Jev」で双腕ロボを動かした結果まとめてみた🦾 3層制御のうち意思決定（2層）をJevに任せ、IKや物理計算はコード側に分離する構造を作りました。500msの応答と0.5円/試行で早くて激安。

Runs on demand
Added
See how it works →X post ↗
X Post · Experiments

Sorting 63,000 emails

WOW 🤯 i just used Jev + @agnt_gg to categorize 63,045 emails in less than 3 minutes for < $1 total runtime: 2 minutes 54 seconds input tokens: 23,430,692 output tokens: 4,528,206 total tokens: 27,958,898 estimated cost: $0.9841 this wou…

Runs on demand
Added
X post ↗
X Post · Games · Experiments

Byte predicts rock-paper-scissors

Can Jev read a human? Play rock paper scissors against Byte. Every round he predicts your next throw before you make it, then plays against that read. His move is committed before you pick, so he can't change it after. Stakes settle over A…

Runs on demand
Added
See how it works →X post ↗
X Post · Games · Experiments

Jev beats Super Mario Bros

Jev beat Super Mario Bros. Every moment, the simulator returns structured data of the game's current state, such as Mario's position, nearby enemies, and terrain, and simulates possible actions like jumping and running. Jev returns a proba…

Runs on demand
Added
See how it works →X post ↗
X Post · Experiments

JSON Render with Jev

An experiment combines json-render with Jev for responsive generated interfaces.

Cadence not verified
Added
X post ↗
X Post · Experiments

An internal-link SEO audit

Jev is WILD for SEO audit 🤯 in 45.1 seconds it read all 586 pages on my site and rebuilt the internal link map. 584 links placed, 139 pages it refused to link because nothing honestly fit. total cost $0.21. Claude Opus 5, same 586 pages, s…

Runs on demand
Added
See how it works →X post ↗
X Post · Experiments

MuJoCo robot-arm control

tested Jev as a real-time robotics policy in MuJoCo it struggled at first, so i split each update into two calls: decide what to do next, then decide how to move the arm and gripper Jev doesn’t accept images, it gets simplified geometry an…

Runs in realtime
Added
See how it works →X post ↗
X Post · Experiments

A predictive app launcher

Also been playing with @typesafeai Jev. Insane. A pile of apps suddenly become possible. What a time to be a builder. Sharing experiments here. First one: Keystroke oracle / predictive launcher A normal launcher ranks by aliases, fuzzy mat…

Runs on demand
Added
X post ↗
X Post · Experiments

Session replay to bug fixes

JEV is INSANE. We gave it 3 million replay events In 40 seconds, it watched 3,247 sessions, caught 132 rage clicks, 116 dead clicks and 95 JavaScript errors, and opened 213 draft PRs to fix them. All for just $2.17. JEV can also flag aband…

Runs on demand
Added
X post ↗
X Post · Experiments

Jev vs. GLM at chess

Jev vs GLM 5.3 at chess! Results: ◾ GLM 5.3 won by checkmate in 29 moves ◾ Jev: ~0.3s and <$0.0001 per move ◾ GLM 5.3: ~5.8s and ~$0.008 per move ◾ The whole game cost 24 cents My main takeaway is that it's often useful to use each one to…

Runs on demand
Added
X post ↗
X Post · Experiments

A Jev query planner

I used Jev to build a query planner! With some tuning, I sped up Postgres queries on the join order benchmark by 12%. Here's what I did

Runs on demand
Added
X post ↗
X Post · Games · Experiments

Jev plays Balatro

Jev from @typesafeai plays @BalatroGame. Every hand it sees the cards, jokers, and blind, then picks play or discard in ~200–500ms. After the ante, it shops too. A full run costs less than a cent.

Runs on demand
Added
See how it works →X post ↗
X Post · Experiments · Agent tooling

Voice intent to tool calls

Played around with @typesafeai Jev today, mostly to understand what it does for tool calling. Instead of an LLM deciding what to do, I substituted that part with Jev. My learning is that we will be able to make the agent act before the use…

Runs on demand
Added
X post ↗
X Post · Experiments

A plain-English logic interpreter

Jev can't reason. So I built an interpreter that reasons for it. It runs a logic language where the facts and rules are plain English.

Runs on demand
Added
X post ↗
X Post · Experiments

Supabase RLS linter

@typesafeai Jev is FREAKING SICK!! Playing around with it to create an experimental RLS Linter for @supabase. You can check it out here: or view the demo 📺

Runs on demand
Added
X post ↗Website ↗
X Post · Experiments

Ego Lite shopping decisions

ego lite + Jev + DeepSeek Flash = stupid fast. ⚡ 3.71s for 20 Amazon product decisions. GPT-5.6 Sol: 54.45s Same task. Same result. 10/10 on both. Less reasoning. Faster decisions. 1× speed.

Runs on demand
Added
X post ↗
X Post · Experiments

Jevals

I wanted an easier way to test my jev requests and see if my results are improving, so I built jevals! It's a local workbench for testng the Noul, Choice, and Score primitives. You can mix requests as well. You can run a seed command to se…

Runs on demand
Added
X post ↗
X Post · Experiments

An always-on assistant with no wake word

Ai is evolving. Jev can be armed at all times. I can speak freely and it knows ( from probabilities ) if im asking my computer to do something or blaberring away at something else. no wake word. speed + affordability + intelligence is gett…

Runs on demand
Added
X post ↗
X Post · Games · Experiments

Connect Four on Omarchy

We put Jev inside OmarchyLinux @dhh to play Connect Four. It reads the board, chooses a move, watches the opponent respond, then Hermes reviews the result and tries again. 6,499 model tokens. Cost: $0.00022113. This is what structured stat…

Runs in realtime
Added
See how it works →X post ↗
X Post · Experiments

A 6502 emulator experiment

Yes yes, Jev can drive, Jev can be a programming language... but lets just skip the middle man ... Jev be the "computer" !!! Motorola 6502 emulation directly on Jev 🤣😅 @typesafeai @dotpem Took a bit "hill-climbing to figure out how to inst…

Runs in realtime
Added
X post ↗
X Post · Experiments

A drone in an asteroid field

Jev flying a drone through an asteroid field. 300ms decisions. Left / right / slow / accelerate. You can watch the decisions happening live on the right. Green = Jev’s own confidence. Purple = “I’m not sure, ask the big model.” 99% of the…

Runs on demand
Added
See how it works →X post ↗
X Post · Experiments

Whose Jev Is It Anyway?

I'm laughing out loud. I turned Jev into an improv judge. Introducing: 'Whose Jev is it Anyway?'

Runs on demand
Added
X post ↗
X Post · Experiments

Danish stock-market backtest

Jev trading the danish stock market for all of 2025(239 trading days). Sentiment-based on market data, news articles, wikipedia &amp; google trends. 8.1 million tokens for $0.32 🤯

Runs on demand
Added
X post ↗
X Post · Experiments

Instant team scheduling with Cal.com

Jev + API = find best overlap among team members in an instant the longest waiting time is our API call round trips 🤯 will come up with ideas how we can bake Jev into our products

Runs on demand
Added
X post ↗Website ↗
X Post · Experiments

Does Jev understand color?

does Jev understand colour? incredible model, very cheap and fast. definitely feels like a leap forward for creating new UX/UI paradigms on top of modern language models.

Runs on demand
Added
X post ↗
X Post · Experiments

Jev Calc

i made a smart calculator notebook using jev it can calculate ANYTHING!! 😅

Runs on demand
Added
X post ↗
X Post · Experiments

A robot arm picks up a ball

I gave Jev a robot arm. Two USB cameras, an SO-101, and @typesafeai's Jev choosing the next move at ~2 decisions a second. After a slip and some debugging, it went back for the ball and got it into the bowl.

Runs in realtime
Added
See how it works →X post ↗
X Post · Experiments

MakerMods robot arm in MuJoCo

Got @typesafeai Jev driving our @makermodsai Metal arm in MuJoCo. Images are of the sim camera is not passed in, sim hands Jev the state of the environment in JSON and it picks the next bounded action (hover, descend, grasp, lift, place) a…

Runs in realtime
Added
See how it works →X post ↗
X Post · Games · Experiments

Chrome Dino with live decision scores

@typesafeai jev playing chrome dino! the model is incredibly fast! - avg response time: ~280ms - spent 116M tokens - cost? just $4.65. - the input here is a snapshot of the chrome runner - jev decides whether to jump, run or duck with a co…

Runs on demand
Added
See how it works →X post ↗
X Post · Experiments

Quality checks for physical-AI action labels

Jev is insane. We are building egocentric training data for physical AI. Jev QA'd 58,643 action labels in under 3 minutes. 90 cents. A friction cost of the Claude &amp; GPT models.

Runs on demand
Added
X post ↗
X Post · Experiments

Intent-aware search in Replicas

small experiment where I used jev to power the global search in replicas. the possibilities of improving UX with models like jev are endless. we can now have things like: - search that understands what users mean - actions that change base…

Runs on demand
Added
See how it works →X post ↗
X Post · Experiments

Keystroke oracle

Also have been playing with @typesafeai Jev, insane! So many immediate use cases and new apps are possible. What a time to be a builder! Sharing some experiments here starting with: Keystroke oracle / predictive launcher: Your launcher ran…

Runs on demand
Added
See how it works →X post ↗
X Post · Experiments

Jev and ChatGPT Task Race

i made a @typesafeai jev+chatgpt instance race against a purely chatgpt instance on two super simple tasks to see the difference, 2-3x faster! :o on first-time tasks chatgpt will write a question set for jev on the fly, and then subsequent…

Runs on demand
Added
See how it works →X post ↗
X Post · Experiments

Jev plays Subway Surfers

jev is insane 🤯 Here is Jev playing subway surfers at super human speed, and also playing 50 games at once. cost less than a cent to do this run. Jev does not replace llms like astra or fable, but opens up an entirely new world of capabili…

Runs on demand
Added
X post ↗
X Post · Games · Experiments

Jev beats Brock in Pokémon FireRed

Jev just beat Brock in Pokémon FireRed in less than 5 minutes. That’s 5 minutes of actual real-world time. You can see the in-game time in the video. This is a completely different way of having an LLM play a game. Jev doesn’t see pixels o…

Cadence not verified
Added
See how it works →X post ↗
X Post · Experiments

Jev Voice Turn Detection

A voice prototype uses Jev to decide whether a speaker has finished and how long to keep listening.

Runs on demand
Added
X post ↗
X Post · Games · Experiments

Jev plays Slay the Spire 2

昨天出的 Jev 模型 适合高频决策，那岂不又是适合打游戏？ 之前用 GPT-6 Astra 代打杀戮尖塔2 能力强但速度慢。 我刚实测用 Jev 打，行动思考只需 0.7秒， 画面我都没看清它就操作完了。 这超人类游戏速度， 着实又让我震惊瘫坐了😅 语言很难描述这感觉，看视频⬇️

Runs on demand
Added
X post ↗Website ↗
X Post · Experiments

Jev Memory Retrieval Experiment

A creator tests Jev in a memory workflow and reports lower token use and faster retrieval; the measurements remain creator-reported.

Runs on demand
Added
X post ↗
X Post · Experiments

Jev $10K Trading Experiment

A creator-posted trading experiment shows Jev making decisions against a $10,000 account while a dashboard tracks the market, account and model output.

Cadence not verified
Added
X post ↗
X Post · Experiments

Five hundred agents in a 3D environment

Jev is actually insane. We benchmarked it, and the results completely change the game for us: > 500 real-time agents > running in parallel > in a 3D environment The preliminary results are crazy: > 500ms average latency > 35 API calls/s >…

Runs in realtime
Added
X post ↗
X Post · Experiments · Games

Jev plays Tetris

Puse a Jev, la nueva IA de @typesafeai, a jugar al Tetris en modo súper difícil. Decidió cada jugada en unos 0,3 segundos. Acomodó 357 piezas e hizo 134 líneas en solo 2 minutos. Increíble.

Runs on demand
Added
See how it works →X post ↗
X Post · Experiments

500 emails for 3.5 cents

Yeah Jev by @typesafeai is very cool. It classified 500 emails in seconds. And it costed 3.5 cents.

Runs on demand
Added
X post ↗
X Post · Experiments

Blitz chess against frontier models

Jev VS Fable 5.1 VS GPT-6 Astra: chess typesafe's new Jev V13 isn't an LLM. it doesn't chat, doesn't explain, doesn't write code — it only makes decisions. so we made it play blitz against frontier LLMs. the test: 5+0 blitz. every move is…

Runs on demand
Added
See how it works →X post ↗
X Post · Experiments · Creative

Jev Sunflower

I was wondering if @typesafeai Jev could generate images so here's what a sunflower looks like according to Jev :

Cadence not verified
Added
X post ↗
X Post · Experiments

Writing one character at a time

i made an llm from first principles with Jev 29 yes/no questions per character: should the next key be a–z, space, comma, or period? highest probability gets append to it, then fed the updated text back in &amp; repeat an autoregressive lo…

Runs on demand
Added
See how it works →X post ↗