User repository on GitHub
Recursive Jev choice over a taxonomy. Select from more than 255 options without breaking TypeSafe Jev's choice cap.
- Forks
- 2
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
Jev directory
CommunityJev Repos & SDKs
Jev Tree — reachjalil
A TypeScript library that walks a taxonomy and calls Jev once per level, so a choice can cover thousands of leaves despite Jev's 255-option limit per question.
- Stars
- 10
- Language
- TypeScript
- License
- MIT
- Reading time
- 3 min
- Published by
- Community
- Author
- reachjalil
- Last verified
- Sep 2026
This is one of several Jev listings for the same project. The canonical entry is Jev Tree — Chuf-H.
About this project
Recursive Jev choice over a taxonomy. TypeSafe’s Jev can only list 255 options in one choice question. Real catalogs are bigger. jev-tree walks a JSON shape and calls Jev once per level (or per partition) so you can select among thousands of leaves. Same stack as jevlogs: typesafe-ai/jev through Vercel AI Gateway, [email protected], Node.js 22+, MIT. Guide · API · Explainer · 20s intro · Dataset.
Quoted from the project's README. Read the full README.
Highlights
- Defaults to 32 options per Jev call and allows maxFanout up to 255; oversized sibling lists are partitioned into labeled buckets.
- Node ids need not be legal Jev choice keys: the library sends o0, o1, and so on, then maps results back to your ids.
- Timeouts and provider failures return reason unavailable instead of inventing a leaf; maxDepth 16 and maxCalls 64 also stop the walk.
- On a 320-leaf incident catalog with 180 labeled tickets, truncating to 255 captured 0% of the tail while the authored tree recovered it.
Quickstart
bash
npm install jev-tree
npx jev-tree # offline demo, no key
npx jev-tree --live --shape taxonomy.json --state "the record"Watch out
MIT-licensed. Live calls need an AI_GATEWAY_API_KEY for Vercel AI Gateway, not an OpenAI key, and the evaluation dataset is synthetic rather than production tickets.
Questions about this entry
Short answers drawn from the linked project or page.
It changes the shape of the question, not the limit. Jev still sees at most maxFanout options per call, 32 by default and up to 255, while jev-tree walks one level at a time and auto-partitions oversized sibling lists into labeled buckets. Cost grows with depth: maxDepth 16 and maxCalls 64 both stop the walk.
No. Timeouts and provider failures return reason: 'unavailable' and the library refuses to invent a leaf. timeoutMs defaults to 8000 per call and timeouts are not retried, so a flaky provider costs you the walk rather than a wrong answer. For tests you can inject a fake evaluator and skip the network entirely.
Your ids do not have to be legal choice keys. The library sends o0, o1 and so on, then maps results back to your ids. Ids still must be unique and __root is reserved. A node with no children, or an empty array, is treated as a leaf, and you can pass one root, a forest, or a shape loaded from JSON.
You still author it. The README says explicit branches beat a 10,000-item flat list, and auto-partitioning is only the fallback for a node with too many children. The published bake-off is synthetic: a 320-leaf catalog with 180 labeled tickets where truncating to 255 caught 0% of the tail. Live calls need AI_GATEWAY_API_KEY, not an OpenAI key.
`npm install jev-tree`, and live calls need AI_GATEWAY_API_KEY, a Vercel AI Gateway key rather than an OpenAI key. There is an offline demo with no key: `npx jev-tree`, or `npx jev-tree --live --shape taxonomy.json --state "the record"`.
Default maxFanout is 32 options per Jev call and can be raised to 255. The per-call timeout is 8000 ms with no retry, maxDepth is 16 and maxCalls is 64; hitting either stops the walk and marks the result unavailable rather than inventing a leaf. validate runs parseShape before walking.
MIT, on the same stack as jevlogs: typesafe-ai/jev through Vercel AI Gateway, [email protected] and Node.js 22+. It is an independent project, not affiliated with TypeSafe or Vercel.
More Jev resources like this
A RubyLLM 2 provider that adds Jev through structured output: build Noul, Choice, and Score questions with a schema builder, pass them to chat.with_schema, and read typed answers from response.parsed.
Jev Repos & SDKs#community#ruby#sdk
Community
kieranklaassen/ruby_llm-typesafeRuby
kieranklaassen/ruby_llm-typesafeRubyA shadcn/ui component library and browser for TypeSafe interfaces: 110 catalog components plus Lab scenes with source previews, install examples, and RTL-aware theming.
Jev Repos & SDKs#community#typescript#react
Community
TypeSafeAI/typesafe-uiTypeScript
TypeSafeAI/typesafe-uiTypeScriptA TypeScript-first toolkit billed as the standard library for System One models: ready-made classify, extract, pickOne, rank and verify methods, a service-agent runtime that runs conversations turn by turn, and a one-message order parser. A Python port is in preview and the project is in alpha.
Jev Repos & SDKs#community#github#typescript
Community
Engid/question-kitTypeScript
Engid/question-kitTypeScriptTypeScript library that wraps Jev in typed decision gates for LLM pipelines. Each gate declares its questions, state, interpret function, and a fail-open or fail-closed policy, returning probability tables your own code thresholds.
Jev Repos & SDKs#typescript#guardrails#sdk
Communitympointer/typesafe-gates
A typed TypeScript client for Jev System One models that sends a state plus Choice, Score or Noul questions to POST /v1/systemone and returns typed answers, with presets for TypeSafe, Zen and custom gateways.
Jev Repos & SDKs#community#github#typescript
Community
sc0d3r/jev-systemoneTypeScript
sc0d3r/jev-systemoneTypeScriptA TypeScript package for one-line local AI decisions: calibrated choice, noul and score answers from open models such as kev-0.6b and kev-4b, with no API keys and no cloud, and models auto-downloaded on first use.
Jev Repos & SDKs#community#github#typescript
Community
amatelic/jevTypeScript
amatelic/jevTypeScriptFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.