Jev Confidence

The Python response models: SystemOneResponse groups answers by id, and records the model name, token usage, and request id.

A 3D robot that understands without generating text: one TypeSafe call per turn with nine typed questions, then code decides whether to act, ask, or shrug.

CLI that fits a per-question confidence threshold to a target accuracy on your own labeled data, verifies it on a held-out split, estimates how much traffic still needs an LLM, and re-checks locked thresholds in CI. It publishes no Jev results of its own.

A Ruby client for TypeSafe's System One API that returns typed Noul, Choice, and Score answers, validates obvious input mistakes before spending a request, and pools HTTP connections across threads.

An independent calibration study of Jev over three public benchmarks and 900 rule-generated support tickets, publishing every raw Gateway response and the quantization limits of returned probabilities.

A community Rails integration built on the typesafe-sdk Ruby gem that adds Rails configuration, persisted usage and cost telemetry, and persistence-backed confidence policies for Choice and Score answers.

A TypeScript library that wraps Jev answers in a policy, a confidence gate, shadow mode, reusable recipes, and an offline eval CLI, so structured decisions can drive actions an application ships.

A toy archipelago air-traffic demo where every in-flight judgment - divert or continue, declare an emergency, give way, and who lands first - is made live by Jev from each plane's local state in about 150 ms.

A Rust derive layer for typed System One decisions: enums and structs become Choice, Score, and Noul questions with compile-time-checked, confidence-gated answers, plus a scriptable fake client for network-free tests.

A dependency-free Swift 6 client for the TypeSafe AI System One API, ported from the official JavaScript and Python SDKs, with Choice, Score, and Noul questions, strict concurrency, and configurable retries.

An open, non-autoregressive System One model with published weights and a TypeSafe-compatible /v1/systemone API; the 395M-parameter OptionMarker checkpoint scores discrete and ordinal questions locally.

An open 0.6B System One decision model that reads unstructured state and typed questions in one forward pass, returning calibrated distributions with zero decoded tokens; weights are published on Hugging Face.

An unofficial PHP client for the TypeSafe AI System One API with typed Noul, Choice, and Score responses, synchronous and Guzzle-promise calls, model discovery, retries, and PSR-3 logging.

An on-demand crypto decision-support terminal that ingests 50-1,000 tweets through a two-tier statistical pipeline, then has Jev's System One answer four typed questions and return a decision card with entry ranges, stop losses, and targets without executing trades.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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%.

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.

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.

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.

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%.

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.

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.

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.

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.

How TypeSafe reports certainty, how it differs from probability, and how to use it architecturally to gate and route decisions.

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.