Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

gojev

Go harness for System One models: hosted Jev, open Kev in-process via llama.cpp, and Cloudflare Clef, with typed helpers and a CLI.

Servicebeta0BSD
View on GitHub ↗

Overview

gojev is a Go harness for System One decision models: TypeSafe's hosted Jev, the open-weight self-hostable Kev, and Cloudflare's Clef on Workers AI. It exposes Bool (probability of yes), Choice (up to 255 options with probabilities) and Score (2 to 10 ordered levels with an interpolated score), returning one answer per question in a single forward pass over a string or JSON state.

The protocol layer lives in the author's forks of the fantasy and catwalk libraries, with providers for Vercel AI Gateway's evaluate endpoint, the native /v1/systemone API or any wire-compatible server, Cloudflare Workers AI including images, and Kev fully in-process through llama.cpp with weights downloaded at runtime. gojev adds catwalk-driven construction, typed decision helpers, Kev conversion tooling and a CLI. It has tests and lint CI.

What the decision model decides

TypeQuestionAnswers
Yes / noBool: yes/no question over stateyes · no
ChoiceChoice: pick one of up to 255 options—
ScoreScore: rate on a 2 to 10 level rubric—

How it works

  1. go get github.com/taigrr/gojev
  2. Pick a provider: Vercel, TypeSafe, Cloudflare or in-process Kev
  3. Ask typed questions with helpers
  4. Use results in your agent or the CLI

Strengths

  • Hosted and fully local options
  • Clef support including images
  • Kev runs in-process

Caveats

  • Depends on the author's library forks

Getting started

go get github.com/taigrr/gojev and configure a provider (TypeSafe, Vercel AI Gateway, Cloudflare or local Kev).

At a glance

Type
Service
Status
beta
Language
Go
Runtime
library + CLI
Has code
yes
Has tests
yes
License
0BSD

Use cases and integrations

How curated lists describe it

  • Go harness for Jev/Kev decision models: TypeSafe, Vercel AI Gateway, and in-process Kev via llama.cpp.

    hellogumbo/awesome-jev

Frequently asked questions

Which models are supported?

Jev, Kev and Cloudflare Clef.

Can it run fully local?

Yes, Kev runs in-process via llama.cpp.

Does it support images?

Via Clef on Cloudflare Workers AI.

Related projects

  • jev

    TypeScript library for one-line local decisions with open models (kev-0.6b, kev-4b, open-jev), no API keys or cloud.

    ProjectbetaTypeScript
  • jev-foundation-models

    Swift 6 bridge that runs System One decision models behind Apple's Foundation Models API, on device with Laya or via hosted endpoints.

    ProjectbetaSwift
  • laya-system-one

    Zero-dependency npm package that runs Laya natively or as WebAssembly in Node, Bun and browsers, drop-in for Jev's /v1/systemone.

    ProjectbetaJavaScript
  • OmniStudio

    Local-first desktop AI workbench for llama.cpp, vLLM and SGLang that includes a JEV typed-decision page and /v1/systemone gateway.

    Servicebeta
  • usejev

    Runs the open LAYA decision model under Bun with native ONNX inference behind a TypeSafe-compatible API, plus an English/Persian playground.

    LibrarybetaTypeScript
  • ollaya★ 1,305

    Rust daemon and CLI that pulls open decision models like Laya and serves them locally behind a TypeSafe-compatible /v1/systemone.

    ServicebetaRust