system-one-models

Evidence-backed use cases, patterns, prompts, and starter code for TypeSafe Jev — a System One model for fast, typed, confidence-aware decisions in software.

The System One harness for System One models. Run Jev and other System One models locally or directly on HarnessRouter.ai.

Fast, typed, calibrated evaluations for LLM and agent outputs, powered by Jev — with simple, framework-agnostic Python APIs

Dynamic latent-state control heads for LLMs: route each query by actual model capability, not task type, to answer, reason, call tools, abstain, or escalate

AX-first computer use for AI agents on macOS with optional Jev (TypeSafe System One) semantic guards: calibrated target/input judgments before an irreversible action, decisions kept in code. Accessibility-tree targeting, window-scoped input, clipboard-safe paste, read-back verification. Ships a pip CLI, a pi package and a DeepSeek Harness plugin.

ReflexBench — open benchmark and evaluation harness for System One models and typed decision engines

Jev-Decision focuses on one thing missing from most current Jev-style open models: a unified decision interface across text and vision.

Code and data for evaluating Jev, a System One model, on scientific decisions and how its choices affect downstream results.

decision-first data cleaning system powered by Jev

Find files by describing them in plain English: find(1) with a semantic --like predicate, answered by TypeSafe.ai's jev model

Reproducible benchmark evaluating TypeSafe AI's Jev (System One paradigm) on Brazil's ENEM 2025 standardized exam. Evaluates typed decision-making, domain-specific accuracy, and RLCD uncertainty calibration against open LLM baselines with an interactive GitHub Pages dashboard.

An abstraction for System One models, currently supporting TypeSafe's Jev and Laya. Typed questions, probabilities, and "don't know" instead of a guess, from Python, the command line, or MCP

[RESEARCH ARTIFACT REPO] Reproducible research on cross-backbone probabilistic decision transfer and learned DecisionCore portability. DOI: 10.5281/zenodo.22947927

Jev-trading is a desktop application that combines the ultra-fast neural network model Jev (by TypeSafe AI) with an intuitive visual interface. We've created a platform where the power of algorithmic trading is accessible without writing a single line of code, messing with Python scripts, or configuring servers in the terminal.

Reliability testing for probabilistic AI decisions in system-one-models.

an Information-Geometric Analysis Toolkit for any System-one (Jev, Jevlike) agent systems

Adaptive Yield: Answer fast, Know when to think, Assign probabilities

A precision-first RAG orchestrator: filters before it generates, verifies before it answers.

Rank your agent skills against a request so your agent can suggest which to invoke; powered by Jev

A spam classifier - using system-one -> system-two architecture.