The raw HTTP contract behind every SDK: POST a state plus typed noul, choice, and score questions to /v1/systemone, and get one answer per question, with error codes and retry guidance.
Sites & Guides·#official#docs#api
Official
The model card for jev-1.13.0: $42 per billion input tokens with free output, a 64k context, 250k tokens per second, and how to list the models your account can call.
Sites & Guides·#official#docs#models
Official
How to shape what Jev evaluates: a string, a named object, or an array of messages. Every question in a request sees the same state and is evaluated independently.
Practices & Patterns·#official#docs#state
Official
The idea behind Jev: models that return typed decisions and probabilities instead of generated text, named after Kahneman's fast-thinking System 1, with confidence to route around uncertainty.
Sites & Guides·#official#docs#architecture
Official
Why TypeSafe trains decision models with RLCD instead of RLHF: calibrated probabilities where 0.2 outcomes happen about 20% of the time, and the case for machine-to-machine automation.
Sites & Guides·#official#docs#evaluation
Official
A maintained list of jev-1.13's known failure modes, literal reading, unreliable counting, dates as text, indirection, context rot, and contradictory criteria, each with a guardrail.
Practices & Patterns·#official#docs#evaluation
Official
Pick one option from a set you define. Returns the option, the full probability distribution, and 0-1 confidence, with up to 255 options and parallel questions that barely add latency.
Practices & Patterns·#official#primitives#choice
Official
Rate content against 2 to 10 ordered levels. Returns a probability-weighted score that can land between levels, per-level probabilities, and confidence, with the arithmetic left to code.
Practices & Patterns·#official#primitives#score
Official
The yes/no primitive: one probability from 0 to 1, where the value is the answer and no separate confidence is needed. True and false criteria pin down subtle boundaries.
Practices & Patterns·#official#primitives#noul
Official
Instructions, option descriptions, score levels, and noul criteria all accept JSON, so you can label question parts, pass schemas and taxonomies, and keep candidate paths alive.
Practices & Patterns·#official#primitives#architecture
Official
Break a judgment into atomic Score questions, normalize each by its top level, and combine them with weights you control in code, as in the resume example's four scored dimensions.
Practices & Patterns·#official#patterns#score
Official
Use the answer to decide what to do and confidence to decide whether to act: the voice-banking example routes below 0.6 to a human and needs 0.85 or more to auto-approve a transfer.
Practices & Patterns·#official#patterns#confidence
Official
Ask everything the system might need in one request, then let code throw away what it does not use. The ticket-triage example sends a category plus four speculative questions in parallel.
Practices & Patterns·#official#patterns#batching
Official
Classify intent with a Choice and complexity with a Score, then send each branch to deterministic code, a specialist model, or a person, with a 0.5 intent-confidence floor.
Practices & Patterns·#official#patterns#routing
Official
The official demo hub, currently home to the Smart Home Assistant, which uses speculative questions to pick a tool and falls back to a general model when nothing fits.
Tools & Integrations·#official#demo#tool-calling
Official
The official SDK hub: Python and JavaScript/TypeScript clients with typed questions and answers plus automatic retries, and the HTTP API for every other language.
Repos & SDKs·#official#docs#sdk
Official
Quickstart for @typesafe-ai/sdk on Node 20+: answer types inferred from your questions, a choice() helper, and one systemOne call to classify a support ticket.
Repos & SDKs·#official#docs#sdk
Official
Usage guide for typesafe-sdk: sync and async clients, uv and pip installs, and one request mixing Noul, Choice, and Score questions with answers grouped by type.
Repos & SDKs·#official#docs#sdk
Official
Semantic search over GitHub's Terms of Service: one request ranks all 218 lines with a Choice while a Noul checks whether the document contains an answer at all, including when it should say no.
Cookbooks & Demos·#official#cookbook#search
Official
A two-stage extraction cascade: a mini model extracts, a Noul battery verifies each field in one request, and a 0.7 gate escalates to a reasoning model, sitting on the cost/quality frontier.
Cookbooks & Demos·#official#cookbook#extraction
Official
Turn natural language into typed function calls: closed-set arguments become Choice questions, optional ones become Noul questions, and 54 questions ship in one request for a trading assistant.
Cookbooks & Demos·#official#cookbook#tool-calling
Official
Decide which of 450 candidate product pairs from two beer catalogues describe the same entity: one three-level Score per pair splits 40 merges, 50 curator reviews, and 360 non-matches.
Cookbooks & Demos·#official#cookbook#entity-matching
Official
Catch wrong and hallucinated citations against RFC 7519 with a string match plus one Choice question: four accurate citations verified at 0.93 or more and all four planted failures caught.
Cookbooks & Demos·#official#cookbook#citations
Official
A repeatability study: an 8-question moderation rubric run 15 times per condition shows a mean probability standard deviation of 0.0098, and a 0.60 uncertainty gate lifts agreement to 99.2%.
Cookbooks & Demos·#official#cookbook#consistency
Official
A 14-Noul claims-triage rubric over one insurance claim, repeated 15 times: mean probability standard deviation of 0.0102 at 111 ms per call, with a 0.30 to 0.70 band for human review.
Cookbooks & Demos·#official#cookbook#consistency
Official
Extract dates by asking seven Choice questions about a date's shape and parts, then resolving them in code: five of six examples auto-accepted and missing dates flagged at 0.46 confidence.
Cookbooks & Demos·#official#cookbook#extraction
Official
Score retrieved passages with four Noul questions each, then route in code: prompt injections dropped, false premises kept as conflicts, and evidence assembled separately for the generator.
Cookbooks & Demos·#official#cookbook#search
Official
Rank 182 agent skills in one request and re-read the top three in a second: over 488 requests, wrong skill loads fell from 16.8% to 7.3% and needless loads from 9.8% to 4.0%.
Cookbooks & Demos·#official#cookbook#agents
Official
Reconstruct Markdown from unformatted text in two requests: 16 Noul line-pair questions stitch split sentences, then 62 questions classify each block's type and companions.
Cookbooks & Demos·#official#cookbook#formatting
Official
Turn wine tasting notes into 67 numeric columns for a CatBoost regressor: five rounds of reading its own errors cut held-out RMSE from 3.09 to 1.77 across 800 unseen reviews.
Cookbooks & Demos·#official#cookbook#research
Official
Regex finds candidate emails, phone numbers, and amounts; Jev picks the requested span so code copies a verbatim value that cannot be invented, with no free-text extraction required.
Cookbooks & Demos·#official#cookbook#extraction
Official
TypeSafe AI's official launch post: the company is out of stealth and developers can get access to Jev from the website, alongside the founder's launch thread.
Sites & Guides·#official#x#launch
Official
TypeSafe's announcement that Jev is available on the Vercel AI Gateway, linking the model page with the Jev API, pricing, and playground.
Tools & Integrations·#official#x#vercel
Official
A Python asyncio runtime that puts Jev in a supervision loop above Codex or OpenCode workers: it judges completion, test sufficiency, and stuck loops, then steers, retries, or stops the job.
Tools & Integrations·#community#python#coding-agent
A staged code-review workflow that uses Jev for bounded judgments over Git diffs or whole codebases, with a local dashboard. Policy and thresholds stay in code while Jev screens five risk areas.
Tools & Integrations·#community#typescript#code-review
An experimental controller that has Jev choose NES inputs to play Super Mario Bros. from structured emulator telemetry instead of screenshots, logging every decision with probabilities and latency.
Cookbooks & Demos·#community#python#games
A web-search app where Jev picks sources, time ranges, and query terms, then ranks results from a dozen engines through Search1API, returning links and relevance scores instead of generated answers.
Tools & Integrations·#community#typescript#search
A PostgreSQL extension that filters, ranks, and classifies table rows with plain-language conditions judged by Jev, batching 20 rows per request and caching answers per session.
Tools & Integrations·#community#postgres#sql
A browser-use agent where Jev picks one action per step from a page's interactive elements. Runs as an MCP server, CLI, or library and returns a step trace, screenshot, and console errors.
Tools & Integrations·#community#typescript#mcp
An MCP server exposing ten Jev judgment tools, including verify, screen, find, rerank, classify, review, and gate, each returning typed probabilities in roughly 150 to 500 ms.
Tools & Integrations·#community#typescript#mcp
Control a headed Chromium window by voice: one Jev request per partial transcript decides intent, target, and completeness in roughly 300 ms, while code applies thresholds and builds the action.
Tools & Integrations·#community#javascript#browser
A single-binary Go MCP server exposing one evaluate tool for noul, choice, and score questions, with one-command registration for Claude Code, Claude Desktop, and Codex.
Tools & Integrations·#community#go#mcp
An agent skill for writing and improving programs that call Jev: question design, state structure, answer composition, confidence thresholds, and how to diagnose a question that keeps coming back wrong.
Practices & Patterns·#community#skills#claude-code
A Claude Code plugin that trims long Bash output after a command runs but before the model sees it: irrelevant chunks are dropped with omission markers while errors, diffs, and code are preserved.
Tools & Integrations·#community#typescript#claude-code
Grep by meaning instead of regex: Jev scores every line, meanings combine with AND, OR, and NOT, and one query matches across Japanese, French, German, Spanish, and more.
Tools & Integrations·#community#javascript#cli
A Pi coding-agent extension with Jev as a decision layer: a gate judges bash, write, and edit calls before they run, and an output judge flags leaked secrets and classifies failures.
Tools & Integrations·#community#typescript#coding-agent
A small TypeScript client for asking typed questions about your data: one ask call mixes yes/no, choice, and score questions and returns probabilities, selected options, and confidences.
Repos & SDKs·#community#typescript#sdk
A Rust CLI that ranks agent skills for the next step with a two-stage Jev pass plus lexical prefiltering, with Claude Code hooks, a TUI, abstention, and --why-not explanations.
Tools & Integrations·#community#rust#cli
A camera-only MuJoCo quadrotor that uses Jev for tactical maneuvers at 2.5 Hz while classical perception and a 50 Hz reflex layer keep safety outside Jev, clearing a five-station course.
Cookbooks & Demos·#community#python#robotics
A pre-alpha PostgreSQL extension for categorical work in SQL: Noul, Choice, and Score functions plus batched variants that classified 38 strings in two requests.
Tools & Integrations·#community#postgres#sql
Render a live Jev meter over any video: Whisper transcribes, every sentence is scored, and ffmpeg exports a 16:9 edit with flags and a scoreboard. Presets cover debates, calls, and pitches.
Tools & Integrations·#community#python#video
Fish-style zsh autosuggestions ranked by Jev: each keystroke asks which of your last 100 distinct history entries you are completing and shows the best match for one-key acceptance.
Tools & Integrations·#community#typescript#zsh
A local decision ledger for Jev-class models: caches decisions keyed by redacted model, schema, and state, replays them deterministically in CI, and publishes bundles without raw state.
Tools & Integrations·#community#rust#caching
A Python walkthrough of typesafe-sdk: sync and async clients, grouping answers into choices, scores, and nouls, and routing uncertain cases to review, with arithmetic kept in Python.
Sites & Guides·#community#tutorial#python
Community
A practical guide through SDK setup, the three primitives, and five patterns including speculative fan-out, confidence-gated routing, composite scoring, and a retrieval-then-judge cascade.
Sites & Guides·#community#tutorial#python
Community
A plain-English guide to Choice, Score, and Noul with Python and curl examples, confidence-gated branching for support, moderation, and fintech, plus a skeptical read of the benchmarks.
Sites & Guides·#community#guide#python
Community
A map of documented Jev use cases, routing, guardrail verification, citation checking, reranking, rubric scoring, bulk classification, and semantic linting in CI, each with a confidence policy.
Practices & Patterns·#community#routing#evaluation
Community
An engineering note that places Jev as a typed micro-decision layer inside an agent harness rather than a planner, with preconditions for confidence gates and fan-out and a runtime split into five parts.
Practices & Patterns·#community#agents#routing
Community
A comparative review of Jev and the open-source Laya encoder: interfaces, deployment, disclosed internals, calibration claims, and the limits of agreement-based workflow evals.
Practices & Patterns·#community#evaluation#calibration
Community
LangChain's langchain-typesafe integration exposes Jev as a TypeSafeClassifier plus experimental model-routing and auto-mode middleware that screens risky tool calls before execution.
Tools & Integrations·#community#integration#python
Community
Diogo Almeida's launch thread: after co-inventing ChatGPT, he spent two years on RLCD and Jev, a System One model he pitches as 20-200x faster and 40-400x cheaper for typed decisions.
Sites & Guides·#community#x#video
Community
A video demo of a bot that asks Jev to decide buy or sell from a live price feed, then places real orders on Kuru's on-chain order book on Monad every 300ms block.
Cookbooks & Demos·#community#x#video
Community
vogel ran Jev over 1,500 of his own emails to test classification quality and posted the results as a video, calling it the most impressive model he has tried for the task.
Cookbooks & Demos·#community#x#video
Community
Kyle Jeong's demo of browser use with Jev and Stagehand: the accessibility tree is the state, actions are the questions, and one remote-browser task cost $0.001 at near-instant speed.
Cookbooks & Demos·#community#x#video
Community
Yuchen Jin's take on Jev for LLM-as-a-judge work: 20-200x faster and 40-400x cheaper, with a screenshot of the model picking which candidate is AGI after a prompt sweep under $0.10.
Practices & Patterns·#community#x#evaluation
Community
Tamara's demo of fast-jev-compaction for Claude Code: instead of summarizing context, Jev scores every tool call and drops the irrelevant ones, with a video of it running.
Tools & Integrations·#community#x#video
Community
Alex Volkov reports that a Jev-powered Claude plugin reviewing unnecessary tool calls cut a session from nearly 1M to 86K tokens in about one second, with the install prompt in the post.
Tools & Integrations·#community#x#claude
Community
Guillermo Rauch says the fx auto-mode safety reviewer, currently on GPT Luna, is up to 18x faster at p95 and more accurate with Jev, and that Jev is coming to Vercel AI Gateway.
Tools & Integrations·#community#x#vercel
Community
OpenRouter's beta announcement for Jev: state plus typed questions returns typed decisions with probabilities, so there is no JSON prompting, parsing layer, or output validation.
Tools & Integrations·#community#x#video
Community
dax previews how fast browser use can be with Jev as the decision layer and OpenCode's browser-use CLI, quote-posting a demo video of the loop in action.
Cookbooks & Demos·#community#x#browser
Community
Gojiberry's Romàn reports scoring 700 high-intent leads and personalized outreach messages in 40 seconds for $0.09, with confidence scores and lead-message mismatch detection.
Cookbooks & Demos·#community#x#video
Community
Vercel CTO Malte Ubl reports that Jev beat an existing classifier eval previously run on Gemini 2.5 Flash Lite, saturating the eval on quality and running 6x faster.
Practices & Patterns·#community#x#evaluation
Community
Browser Use's Ultrafast agent puts Jev in charge of picking each browser action from a dynamic DOM action space. The author reports a flight search in 7 seconds for about $0.0039.
Tools & Integrations·#community#x#video
Community
A hype-free framing of Jev: not a replacement for GPT or Claude, but a new tool class for fast, cheap structured decisions, like 2016-era classifiers with 2026-level intelligence.
Sites & Guides·#community#x#explainer
Community
A 33-second demo clip captioned "when a designer gets access to Jev", among the most-liked organic Jev posts, showing a polished interface built on top of typed decisions.
Cookbooks & Demos·#community#x#video
Community
A curated thread of the strongest Jev community projects found on X within the first days of launch, from the creator behind the Jev voice browser and a full video tutorial.
Sites & Guides·#community#x#roundup
Community
A 22-minute walkthrough covering what Jev is, API setup, and three demos: a voice-controlled browser, AI memory, and a YouTube predictor, with chapter timestamps in the post.
Sites & Guides·#community#x#video
Community
The pg-jev author demos jev(), a PostgreSQL extension that filters table rows with natural-language conditions and no index or embeddings, judging 129 rows in about a second.
Tools & Integrations·#community#x#video
Community
A widely read counterpoint to Jev-based context compaction: compaction is reconstruction, not filtering, and scoring tool calls to drop history misunderstands context management.
Practices & Patterns·#community#x#compaction
Community
The WebMCP benchmark author reports that Jev paired with the small Mercury 2.5 model solved 100% of tasks at roughly 112x lower model cost than GPT-6 Astra with computer use.
Practices & Patterns·#community#x#benchmark
Community
A short demo of a model router built with Jev: each request is classified by Jev and forwarded to the model judged most suitable, shown as a screen recording.
Tools & Integrations·#community#x#video
Community
The jevmeter author runs five yes/no questions per sentence across a debate: 1,191 Jev calls, 1.18M tokens, 415 ms median latency, and $0.0497 for the full run.
Cookbooks & Demos·#community#x#video
Community
A DuckDB extension that classifies rows in CSV, Parquet, or DuckDB tables with Jev, reported at about 10 seconds for 1,000 rows and framed as more ergonomic than a bespoke classifier.
Tools & Integrations·#community#x#duckdb
Community
Cloudflare's developer account announced that Jev is available on AI Gateway, callable from Workers with typed Noul, Choice, and Score questions returning structured answers.
Tools & Integrations·#community#x#cloudflare
Community
An independent Jev-compatible public API backed by Qwen3.6-35B-A3B and SGLang radix caching, reporting 64 parallel typed-decision tasks in under a second.
Repos & SDKs·#community#x#video
Community
End-to-end guide to a ticket-routing workflow with confidence thresholds, zero-data-retention options, and mock-based tests that need no network call.
Sites & Guides·#community#vercel#tutorial
Community
Use Jev as a Braintrust judge scorer, then inspect selected answers, confidence, and probabilities alongside the scorer answer. Includes tracing for your own calls.
Tools & Integrations·#community#evaluation#observability
Community
An agent-first CLI for System One: ask, rank, extract, verify, screen, find, and decide from the shell, with an installable agent skill. Published on PyPI.
Tools & Integrations·#community#cli#agent
@jev.fn compiles a Python function definition (parameters, docstring, return annotation) into a state plus typed questions, then validates the answers back into Pydantic models.
Tools & Integrations·#community#python#decorator
Community
An unofficial Go client for the System One API: typed judgments and calibrated probabilities instead of generated text.
Repos & SDKs·#community#sdk#go
Community .NET SDK for the System One API: typed noul, choice, and score questions with structured, confidence-scored answers.
Repos & SDKs·#community#sdk#csharp
A community PHP and Laravel SDK for the Jev model series, wrapping typed questions and probability answers for the PHP ecosystem.
Repos & SDKs·#community#sdk#php
Eight minimal, zero-dependency examples applying Jev to mechanical and electrical engineering: CAD/CAE/CAM routing, FEM result triage, DFM screening, and BOM alignment.
Repos & SDKs·#community#engineering#python
A community playground with over a hundred use cases, games, and model challenges, with editable prompts and side-by-side comparisons.
Tools & Integrations·#community#playground#nextjs
A monorepo of roughly 20 latency-focused demos: moderation firehoses, rerankers, inbox triage, macOS launchers, and closed-loop control simulators, each with p50/p95 telemetry.
Repos & SDKs·#community#demos#typescript
One AI trade decision every Monad block: Jev evaluates the market state and returns a typed decision the bot acts on, on Kuru's MON-USDC market.
Repos & SDKs·#community#trading#realtime
Train a small model that chooses among a changing list of text options, one probability per option in a single pass. Includes Doom, chess, and Wikispeedia demos.
Repos & SDKs·#community#training#research
Run a Pydantic AI agent on Jev: each field of the output type becomes one question, the prompt is the state, and confidence comes back per field.
Tools & Integrations·#community#pydantic#python
Community
A practical tour with JavaScript examples: typed choices, scores, probabilities, real latencies and costs, the jagged edges, and how to split jobs correctly.
Sites & Guides·#community#tutorial#javascript
Community
Give Jev a screen snapshot and a task, and it chooses the next action for agent-driven mobile QA. Ships as a CLI you can point at a real app.
Sites & Guides·#community#mobile#qa
Community
An MCP server that puts Jev in the coding loop for Cursor, Codex, and other MCP clients: coding-loop checks, ranking, verification, and review tools.
Tools & Integrations·#community#mcp#coding-agent
An open Python toolkit with question builders, confidence gates, tiered routing, a concurrent corpus evaluator, calibration tooling, a CLI, and an MCP server, with published benchmarks.
Repos & SDKs·#community#python#mcp
Community
Jev-backed auto mode for the Pi coding agent: semantically approves bash, write, and edit tool calls, and fails closed whenever a decision cannot be made.
Tools & Integrations·#community#coding-agent#permissions
The launch story: why TypeSafe built decision models, the side-by-side demo, the Doom and Wikiracing bots, and the published workflow evals.
Sites & Guides·#official#announcement#benchmarks
Official
AI Gateway's evaluation modality: call Jev as `typesafe-ai/jev` through the AI SDK's experimental_evaluate API, with structured answers instead of free-form text.
Tools & Integrations·#community#vercel#integration
Community
Model card for `typesafe-ai/jev`: evaluation type, $0.042 per 1M input tokens, zero output token charge, and a hosted playground.
Tools & Integrations·#community#vercel#pricing
Community
Access Jev through OpenRouter's decisions endpoint, an alternative route to early access that still uses TypeSafe's question and answer shapes.
Tools & Integrations·#community#openrouter#api
Community
The canonical introduction: Jev is TypeSafe's flagship model and the first System One model. Send state and typed questions, get structured answers your code can use directly.
Sites & Guides·#docs#official#primitives
Official
Everything you need to make your first request: install an SDK, get an API key, and send state plus typed questions.
Sites & Guides·#docs#official#quickstart
Official
The three TypeSafe question types, the typed answers they return, how to choose between them, and how to ask several in a single call.
Practices & Patterns·#official#primitives#choice
Official
How TypeSafe reports certainty, how it differs from probability, and how to use it architecturally to gate and route decisions.
Practices & Patterns·#official#confidence#routing
Official
Architectural patterns for System One systems: speculative fan-out, confidence-gated routing, composite scoring, and intent routing.
Practices & Patterns·#official#patterns#architecture
Official
Design AI-powered software by keeping code in control and giving System One narrow, structured decisions to make.
Practices & Patterns·#official#patterns#architecture
Official
Explore TypeSafe use cases by industry and turn promising ideas into concrete software workflows.
Practices & Patterns·#official#use-cases
Official
A 13-question regulatory briefing that shows batching every question into one call is 12.2x cheaper and 10x faster with no change in answers.
Cookbooks & Demos·#official#cookbook#batching
Official
One question per query-candidate pair raises top-1 accuracy from 5% to 18% and top-10 accuracy from 38% to 62% on 40 CLERC legal queries.
Cookbooks & Demos·#official#cookbook#reranking
Official
Screen every message going into and out of an LLM app: a Noul describes possible hazards, a Score rates severity, and code thresholds the probabilities.
Cookbooks & Demos·#official#cookbook#guardrails
Official
Classify documents through deep patent, retail product, biomedical, and source-code hierarchies using parallel beam search over Choice probabilities.
Cookbooks & Demos·#official#cookbook#classification
Official
Classify SEC annual reports into 75 industry groups with one Choice each, then read confidence to decide when to fall back to a broader division.
Cookbooks & Demos·#official#cookbook#classification
Official
A smart home assistant that uses TypeSafe to evaluate user requests, pick a tool, and fill its arguments from plain language.
Cookbooks & Demos·#official#demo#tool-calling
Official
Drop-in skill for Claude Code, Codex, and other agent environments that teaches the right way to structure states and questions.
Tools & Integrations·#official#agent#skill
Official TypeScript/JavaScript client with inferred answer types, retries, and a question DSL. `npm install @typesafe-ai/sdk`.
Repos & SDKs·#official#sdk#typescript
Official sync and async Python client for the TypeSafe API with typed questions, retry policies, and inferred answers. `pip install typesafe-sdk`.
Repos & SDKs·#official#sdk#python
Official drop-in TypeSafeClient replacement backed by OpenAI, Anthropic, and compatible LLM APIs, for comparing Jev against chat models on the same questions.
Repos & SDKs·#official#adapter#python
Get an API key, manage models, and try the playground: three walkthrough lessons plus realistic use cases like resume screening and ticket routing.
Tools & Integrations·#official#playground#api-keys
Official
