← Back to live feed

Thursday, Sep 17, 2026

1
TypeSafe AI Launches Jev Model on OpenRouter With Free Output Tokens and 190x Faster ResponsePVT:OPRT
topics πŸ€– AI tags AIAI ModelsAI ReleasesAI InfraAI Inference PVT:OPRT keywords

Jev returns typed decisions and probabilities instead of generating text tokens in a new beta release from TypeSafe AI on OpenRouter. The System One model eliminates JSON prompting and parsing layers by processing an application's state and a typed question to produce a direct decision. Users can query the system for binary Noul yes or no answers, a specific Choice from a defined set, or a Score on a custom scale.

The model costs $0.042 per million input tokens with free output tokens and provides end to end response times between 70 and 500 ms. TypeSafe AI reports performance gains of up to 190x faster speeds and 440x lower costs compared to traditional large language models on published workflows. Primary applications include routing support tickets, flagging fraudulent signups, and determining if automated workflows require human intervention.

Image via @fazxes on X
You're reading an older version of the story.
Bespoke Labs Releases Nimble Open Model That Hits 90% on Jev Eval
21 tweets β€’ 13 sources
Continues from Thursday, Sep 17
TypeSafe AI Clears 140,000 Waitlist Users for First System One Model in 36 Hours
23 tweets β€’ 20 sources
See all 31 tweets β†’