Intent detection with decision models

Rails engine for voice and typed commands: define actions in a Ruby DSL and Jev picks the matching command or page control.

Chat copilot for Mac, Windows, Android and an iOS keyboard that reads chats, judges with Jev or DeepSeek and drafts replies.

Pre-registered independent evaluation of Jev on intent classification and confidence-gated cascades, with a transparent erratum.

Benchmark harness pitting Jev against Claude Sonnet 5 and Haiku 4.5 on Banking77 intent sorting with a confidence gate.

Benchmark of Jev against low-cost LLMs on 130,000 classification decisions across five datasets for accuracy, cost and latency.

OEV: 184M-parameter decision engine that scores options in one pass, reporting 0.7705 on typed-decisions and 22 ms per question.

Production control plane for OpenClaw agents: Jev routes each Slack message to its task, with Temporal workflows, an evaluator and memory.

npm library for chess with Jev: natural-language move intents to legal moves, parallel move evaluation and historical persona opponents.

Java SDK that lets natural language invoke tinystruct @Action methods, using Jev as a semantic dispatcher with confidence gates.

I Ching divination site where charts and verdicts come from code and Jev only classifies the question's topic and screens unsafe ones.

AstrBot group-chat plugin fork that adds a 50-70 ms System One check on whether the bot should chime in, after the original random gate.

Zero-build local playground for Jev with browser-agent, intent routing and context-pruning presets, including a Playwright agent.

Native macOS launcher and window manager that optionally uses Jev to match loose natural-language commands to known actions.

Draw on a tldraw canvas by voice while pointing a webcam-tracked finger; Jev decides command, action, shape, colour, target and place.

Symfony demo that triages French public-service requests with one Jev call each: intent choice, urgency score and bug noul.

E-commerce support sandbox where Jev classifies intent, emotion and escalation and blocks risky agent replies before sending.

Customer-service chatbot that asks every level of a routing tree in one Jev call and hands off on human requests or low confidence.

Collection of 22 latency-focused Jev demo apps built by Devin, from shell guards and log sentinels to search and voice turn-taking.

Semantic command palette demo where one Jev Choice over 77 commands per keystroke gives the ranking, beside a fuzzy-match baseline.

Chat interface with no generative model: Jev picks intent, stance, claim and tone, and a deterministic compiler writes the reply.

Map where you ask anything in plain words and Jev gives every nearby place a probability, lighting up those likely to match.

One text box that morphs into the right UI card as you type, with Jev answering 14 typed questions per input and code doing the maths.

Fastino's 340M Apache-2.0 decision classifier that scores caller-defined labels in one forward pass for routing, intent and scoring.

Open multilingual System 1 decision engine built on Laya checkpoints with a router that picks the best checkpoint per request.

NIRNAY 450M: open Apache-2.0 decision model fine-tuned from Laya that reports 87.9% on Banking77 with fitted ECE 0.045.

Research router where a Qwen3-1.7B belief model scores intent candidates and a small policy decides to commit, retrieve more or hand off.

nev-lite: 11.8M-parameter static-embedding decision head that scores fixed option sets with no transformer pass at inference.

Independent benchmark of Jev, constrained-output LLMs and baselines on intent and spam tasks: accuracy, latency, cost, calibration.

Android chat copilot: Jev judges intent, needs and risk in visible chats, then ranks drafted replies to fill, never send.

Page companion where a site's actions are WebMCP tools and Jev picks the action a visitor's request means, asking back when unsure.