Jev launches: a System One model for decisions
TypeSafe founder Diogo Almeida introduces Jev, its RLCD training method, and its focus on software decisions. The post includes a 2:56 launch video.

Developed by former OpenAI researchers, Jev doesn't generate text. Instead, it returns one of the given options with a probability. For tasks like classification, it claims to be up to 200x faster than major AI. Early access has begun.
文章を書かないAIが出てきた。元OpenAIの研究者たちが作った「Jev」は、答えを文章で書かず、渡された選択肢から一つを確率つきで返す。分類のような仕事なら、大手のAIより最大200倍速いという。早期アクセスが始まった👇