Models and aliases
TypeSafe docs page listing Jev version IDs, aliases, request limits, rate limits and pricing for /v1/systemone.
This summary is based on how curated lists describe the project; its README was not available to review.
Overview
This is the TypeSafe documentation page for Jev models and aliases. According to the list summary, the endpoint is POST https://api.typesafe.ai/v1/systemone, the current version is jev-1.13.0, and the aliases jev-latest and jev-preview distinguish stable and preview releases.
The summary also records limits: 64k tokens per request, with state plus the longest question up to 32k; rate limits of 250k tokens per second and 1,200 requests per minute with dynamic adjustment; text input only; Choice up to 255 options and Score with 2-10 tiers. There is no fine-tuning, all accounts share the same weights, customer data is not used for training, and accuracy is best in English. Developers integrating Jev should check this page before pinning versions or sizing workloads.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Choice question with up to 255 options | — |
| Score | Score question with 2-10 tiers | — |
How it works
- Choose a pinned version (jev-1.13.0) or an alias.
- Keep requests under 64k tokens and state plus question under 32k.
- Stay within rate limits, which adjust dynamically.
Strengths
- Clear limits for capacity planning.
- Stable versus preview aliases support safe upgrades.
- States customer data is not used for training.
Caveats
- Text input only; no fine-tuning.
- Limits and pricing may change; check the live page.
Getting started
Send requests to POST https://api.typesafe.ai/v1/systemone with model jev-1.13.0 or an alias.At a glance
- Type
- Article
- Status
- production
- Runtime
- hosted API
- Has code
- no
- Has tests
- no
Use cases and integrations
How curated lists describe it
Endpoint `POST https://api.typesafe.ai/v1/systemone`; `jev-1.13.0`, aliases `jev-latest`/`jev-preview`; 64k/request, state + maximum question length 32k; rate limits of 250k tok/s and 1200 req/min with **dynamic adjustment**; text input only; Choice≤255, Score 2–10 tiers; **no fine-tuning, all accounts share the same weights**; customer data is not used for training; most accurate in English.
wh000wh000/awesome-jev-liveVersion IDs, current limits, pricing, and the distinction between stable and preview aliases.
cobanov/awesome-jev
Frequently asked questions
What is the current Jev version?
The list summary names jev-1.13.0, with aliases jev-latest and jev-preview.
Can I fine-tune Jev?
No; all accounts share the same weights.
How many options can a Choice have?
Up to 255; Score supports 2-10 tiers.
Related projects
Server behind api.typesafe.pro, a free anonymous System One-compatible endpoint that forwards to TypeSafe on the operator's key.
ServiceproductionBrowser form for building Jev requests from pasted text without hand-writing JSON, then rerunning to see how steady the answers are.
ProjectbetaJavaScriptTerminal REPL for shaping System One requests before writing code, with lessons, a sketch editor, conversation turns and offline simulation.
ServicebetaTypeScriptNushell module for TypeSafe's System One API: pipe content in, ask noul, choice and score questions, and filter on the numbers.
ServicebetaNushellOne-page playground and test bench for Jev with TypeSafe, OpenRouter, LLM-baseline and mock backends behind one /v1/systemone contract.
ServiceprototypeTypeScriptStatic browser interface for making structured Jev decisions through the Pollinations API, with multilingual READMEs.
ServiceprototypeJavaScript