jev-lab
Small Node TypeScript CLI for trying Jev on synthetic support text: one question, three in one request, and latency benchmarks.
Overview
Jev lab (menny1337) is a small Node.js TypeScript command-line tool for trying Jev on synthetic support text. You can run one urgency question, ask three questions in one request, or compare request latencies with configurable warmups, iterations and timeout. There is no web server or UI.
It pins the official TypeScript SDK at v0.6.0 via a GitHub release tarball with an integrity hash. The README is written for Windows PowerShell and shows how to set the API key without putting it in command history, and the CLI refuses to send requests with an absent or placeholder key. Help and tests need no key.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Score | How urgent is this support message? | — |
How it works
- npm ci, test, typecheck and build
- Set TYPESAFE_API_KEY securely
- Run smoke, batch or benchmark
- Compare latencies
Strengths
- Pinned official SDK
- Careful key handling
- Simple latency benchmark
Caveats
- Synthetic text only
- Windows-focused instructions
Getting started
`npm ci && npm test && npm run build`, set TYPESAFE_API_KEY, then `npm run smoke`, `npm run batch` or `npm run benchmark`.At a glance
- Type
- Benchmark
- Status
- prototype
- Language
- TypeScript
- Runtime
- Node 20+ CLI
- Has code
- yes
- Has tests
- yes
Use cases and integrations
How curated lists describe it
TypeScript experiments, evaluations, and latency benchmarks for TypeSafe's Jev model.
hellogumbo/awesome-jev
Frequently asked questions
What does it test?
An urgency question, a three-question batch, and request latency.
Which SDK version?
The official TypeScript SDK pinned at v0.6.0.
Do tests need a key?
No, help and tests run without one.
Related projects
Rust port of TypeSafe's system-one-adapter that answers System One questions with OpenAI or Anthropic LLMs for comparison with Jev.
ProjectbetaRustTypeSafe's official drop-in replacement for the SDK's system_one call backed by OpenAI, Anthropic or Gemini, for comparing Jev with LLMs.
LibraryproductionPythonC# System One client and benchmark comparing Microsoft-Decision-1, Jev and open-jev on the same labelled support tickets.
LibraryprototypeC#Small harness calling Jev via OpenRouter's Decisions API, with 17 doc-derived cases and a latency and cost benchmark against two LLMs.
BenchmarkprototypePythonOne-page playground and test bench for Jev with TypeSafe, OpenRouter, LLM-baseline and mock backends behind one /v1/systemone contract.
ServiceprototypeTypeScriptRust client, CLI, terminal UI and eval harness for asking System One questions of a local model served on Xeon Phi cards.
LibraryresearchRust