Typesafe · Decision
Jev 1.13.0
Newer version: Jev 1.14.0. Our catalogue has no release date for Jev 1.13.0. We compared the version names in our catalogue.
Jev from TypeSafe is a System One evaluation model: instead of generating prose it answers typed questions about a piece of content. Send a state plus a map of questions and get back structured answers with full probability distributions - Noul for yes/no, Choice for picking one option from a set, and Score for rating against an ordered rubric. Every request is charged on input tokens only; output tokens are free.
Evidence and sources
All rankings:
#11 on OpenRouter this month
#14 on OpenRouter today
#15 on OpenRouter this weekSources: TypeSafe Models Docs data docs, Vercel AI Gateway data, NetMind.AIPrimary data from TypeSafe Models Docs
from Vercel AI Gateway: name
from NetMind.AI: description
Advantages and trade-offs
Compared with 10 representative decision models
Advantages
Advertised function calling support
Function calling advertised by 2 of the 5 peers that list their features.
Input & output
Inputs as stated by providers; output from the model’s typeText
Images
Choices and scores
Also known as: jev-1.13, jev-1.13-free, typesafe-ai/jev-1.13.0, typesafe/jev-1.13, typesafe/jev-1.13.0, vercel_ai_gateway/typesafe-ai/jev-1.13.0
Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works