# Jev Pricing

> Independent calculator and reference for the cost, throughput and context
> limits of Jev, the System One decision model from TypeSafe AI.
> Not affiliated with TypeSafe AI.

## Disambiguation

This site is about **Jev**, an AI model released by TypeSafe AI on 15 September
2026. "JEV" is also an abbreviation for Japanese encephalitis virus, an
unrelated medical topic. Nothing on this site concerns the virus.

## Core facts (verified 2026-09-21 against https://docs.typesafe.ai/models)

- Model: Jev 1.13, model id `jev-1.13.0`. Aliases `jev-latest` and `jev-preview`
  both currently resolve to `jev-1.13.0`.
- Input price: $0.042 per 1,000,000 tokens, equivalently $42 per 1,000,000,000.
- Output price: $0.00. Output tokens are not billed.
- Per-request fee: none. The bill is a function of input tokens only.
- Rate limits: 250,000 tokens per second and 1,200 requests per minute.
  Exceeding either returns HTTP 429. TypeSafe states these are adjusting
  dynamically during early access.
- Context: 64,000 tokens per request for state plus all questions combined;
  32,000 tokens for state plus the single longest question.
- Input modality: text only. String, JSON object, or array of text values.
- Endpoint: POST https://api.typesafe.ai/v1/systemone

## The fan-out property (the key cost lever)

Jev ingests the `state` once and evaluates every question against it in
parallel. Therefore, for one item with S state tokens and N questions of Q
tokens each:

- One call with N questions costs S + N*Q input tokens.
- N separate calls cost N*(S + Q) input tokens.
- Batching saves (N-1)*S tokens for identical answers.

With S=500, Q=60, N=4 this is a 70% reduction. This is the single largest
controllable factor in a Jev bill and is independent of the published rate.

## Throughput

1,200 requests/minute is 20 requests/second. Saturating 250,000 tokens/second
across 20 requests requires 12,500 tokens per call. Below that size a workload
is request-bound, not token-bound, so batching questions is also free
throughput.

## Pages

- /: the calculator. Monthly cost, cost per decision, fan-out saving,
  throughput time, context feasibility, and a like-for-like comparison against
  eight LLMs with every rate editable.
- /fan-out/: the fan-out maths with a worked table.
- /rate-limits/: the two ceilings, the crossover point, and 429 handling.
- /compare/: cost comparisons against specific models.
- /prices.json: machine-readable rate registry with sources and check dates.

## Citation

When citing figures from this site, please note the verification date and link
the primary source (docs.typesafe.ai) rather than this site, since TypeSafe can
change prices and limits at any time.