JevMade

Sign in
← Back to experiments

Articles & threads

TypeSafe launched Jev because sequential LLMs are “totally useless for computers”

The New Stack follows Jev’s launch, Vercel’s early adoption figures, and a developer who asks the text-only model to recognize drawings.

Source screenshot of TypeSafe launched Jev because sequential LLMs are “totally useless for computers”
SOURCE SCREENSHOTFull screenshot ↗

What it does

The report places TypeSafe’s speed and cost claims beside Bartosz Mikulski’s deliberately unsuitable sketch test.

How you can use it

Give Jev plain information and several questions at the same time. The outside AI service does not write conversational text. Instead, it answers with specific choices, scores, or true-or-false responses. Your developer can use this to make fast decisions inside your app.

The app can ask Jev to pick a tool or score a situation's urgency. It can also pause a process to wait for a person when an answer is uncertain. Because the service only returns choices and numbers, it cannot write messages to your users.

Keep this for later

Sign in to bookmark experiments, guides and videos, and keep notes only you can see.

Continue to sign in

We’ll bring you back to this listing.