ModelsSep 17, 2026 · 18:528 sources

TypeSafe AI Clears 140,000 Waitlist Users for First System One Model in 36 Hours

From huggingnews.com

Image: huggingnews.com
AI;DR

TypeSafe AI released the Jev AI model, which uses a non-generative architecture to answer structured questions in parallel and provides typed choices and confidence scores. The company granted access to 140,000 waitlisted users in under 36 hours, with the system available through the Vercel AI Gateway. Deployment data show a 71% reduction in costs and a 90% reduction in completion time for a production dispatch process, and tests reported Jev performing 20 to 200 times faster and 40 to 400 times cheaper than text-generating models.

Read the sourceSign in to suggest

Sources

+ 5 more
  1. SUPPORT@kunchenguid“the saving from Jev still resulted in a -71% reduction in cost and -90% reduction in wall time of completing the whole dispatching process”x.com
  2. SUPPORT@completeskeptic“I used Jev to classify 1,018 AI research papers. The result: $0.08 total cost and 256ms median end-to-end latency per paper”x.com
  3. SUPPORT@completeskeptic“results of using jev on one of our security pipelines. we already have to use smaller (and dumber) models”x.com
  4. SUPPORT@rauchg“Jev is up to 18x faster (p95) *and* more accurate. It's coming to @vercel AI Gateway and likely new default.”x.com
  5. SUPPORT@rauchg“Jev is now available on the @vercel AI Gateway”x.com
#llm#inference#agent#open-source#fine-tuning#benchmark#typesafe-ai#waitlistMarkdown
Content history (1)
  • 2026-10-04 · Summary · vi · Wording fix
    Mô hình AI mới Jev của TypeSafe AI sử dụng kiến trúc phi sinh (non-generative), trả lời c…
    Mô hình AI mới Jev của TypeSafe AI sử dụng kiến trúc phi sinh, trả lời các câu hỏi có cấu…

One ranked edition, every morning

The stories that matter today. Free, cancel anytime.

Telegram @aidr_today