Calibration & Research

Mangaba AI's local, calibrated typed-decision model for Portuguese text, trained to optimize the Brier score; proprietary since v5.

TypeLLM: Jev-inspired type-safe decoding for existing autoregressive LLMs on SGLang, with schema-guaranteed typed outputs.

Open one-pass typed-decision model with a dependency-free C99 runtime that also builds to a 66 KB WebAssembly module.

Local open alternative to Jev using a 1.58-bit ternary 8B model to answer booleans, choices, scores and Pydantic extraction in ~25 ms.

Free hosted Jev-style multimodal decision model API (about 20B active) that picks actions from images and options, aimed at robot control.

Runs the open Laya decision model entirely in a browser tab with WebGPU and ONNX Runtime Web: no server, no API key.

ComfyUI nodes for Jev-style noul, choice and score decisions over images and text using a local imajev adapter on Qwen3.5.

llama.cpp-based C++ server that turns any LLM into a System One decision model, reading answers from next-token logits.

Nautilus Compass agent memory layer whose jev-trust middleware logs Jev decisions and measures calibration in your domain.

Optimises the instructions and criterion descriptions of a Jev Choice from labelled data with EvoPrompt or GEPA, scored by Brier.

Decision-forensics library that explains a Jev answer by ablating parts of the input and measuring how far the probability moves.

Three open 4B decision models for English, Turkish and German that serve the Jev /v1/systemone API on a CPU in about 3 GB of RAM.

Japanese System One decision model on ModernBERT-ja (314.6M) returning typed choice, score and bool distributions with no generation.

Adapter that turns existing LLMs into Jev-style decision functions with option scores, optional calibration and review thresholds.

Decision sidecar and Copilot CLI pre-turn hook that mimics Jev choice, noul and score using Azure Foundry Cohere Rerank.

Toy chat that makes Jev talk by repeatedly choosing its next word from a tiny caveman vocabulary.

Python library that imitates Jev-style typed decisions (choice, score, yes/no) on local LLMs via Ollama, vLLM or llama.cpp.

MLX library that turns an existing local model into a decision engine by reading candidate logits, with no fine-tuning, on Apple Silicon.

Local browser app that turns Jev's per-pixel colour probability distributions into textured paintings.

Creative npm package that makes UUID v4-shaped strings from Jev's probabilities on prompt variations; not cryptographically random.

Experiment running a trained jevlike option scorer as ESP-IDF firmware on an ESP32 microcontroller.

Browser-only Jev-style decisions in Chinese or English using a zero-shot NLI model on ONNX Runtime Web, with honest limits.

Turns any local LLM into a Jev-style engine via masked-logit softmax, or proxies OpenAI-compatible APIs or Jev; includes a chat crush radar.

Python library that turns an open LLM into a Jev-style classifier via constrained single-token choice indexes in one forward pass.

Two-tier macOS decision engine: a MiniLM encoder on the Neural Engine decides, escalating unsure cases to Apple's on-device model.

Interactive web workbench for exploring and benchmarking Jev's Choice, Noul and Score via OpenRouter, with threshold transfer curves.

Open Jev-style decision model built on Gemma 3 270M with a scoring head, scoring all options in one forward pass on CPU.

Self-hosted encoder model giving Jev-compatible choice, score and noul answers in one pass, pitched as a better-calibrated Laya successor.

33-chapter Chinese tutorial on agentic RL and GRPO, including a chapter on when Jev-style external decision models beat training with RL.

Joke decision model trained to answer wrongly with total confidence: 1.9% accuracy at 0.957 mean confidence, via a Jev-style API.