CITATION — REFERENCE ENTRY

Models — TypeSafe AI documentation

Revision 0e7de9dd-db42-46a9-a4b4-3830a397efff · 11/5/2026, 9:54:55 PM UTC

Key: typesafe2026models
Authors: TypeSafe AI
Issued: 2026
Type: webpage
Container: TypeSafe AI documentation
Publisher: TypeSafe AI
URL: https://docs.typesafe.ai/models

Raw CSL JSON

{
  "URL": "https://docs.typesafe.ai/models",
  "type": "webpage",
  "title": "Models",
  "author": [
    {
      "literal": "TypeSafe AI"
    }
  ],
  "issued": {
    "date-parts": [
      [
        2026
      ]
    ]
  },
  "accessed": {
    "date-parts": [
      [
        2026,
        11,
        5
      ]
    ]
  },
  "language": "en",
  "publisher": "TypeSafe AI",
  "container-title": "TypeSafe AI documentation"
}

Claims

limits

By default a request may hold 64,000 tokens, and the state plus any one question 32,000 tokens. Jev is not fine-tuned on customer data; it is customised only through the state and the wording of questions.
pricing

Jev costs $0.042 per million input tokens; output tokens are free.

"You pay per input token. Output tokens are free."

Quote language: en
v1-14

Version jev-1.14.0, available from October 14, 2026, added a fourth question type, Box; version 1.13.0 supports Choice, Score and Noul. Both listed versions accept text and PNG or JPEG images, but not audio or video. The documentation gives no release date for version 1.13.

Available in

en - English