AIToday
Large Language ModelsOpen-Source AIAI Business & IndustryTechCrunch AIPublished: Oct 17, 2026, 01:00 JST

Jev hits 35% of paid teams in two weeks, Vercel says

Jev hits 35% of paid teams in two weeks, Vercel says

3 Key Points

  1. What happened

    Vercel data published Thursday shows TypeSafe AI's Jev, a lightweight decision model for choosing among predetermined outputs, was used by 35% of paid teams within two weeks of its September 15 limited release — nearly double Sonnet 5.5's reach.

  2. Why it matters

    The gateway figures suggest the buzz around Jev translated into real usage, and Vercel itself was an early adopter, replacing GPT-6 Luna with Jev for classifying harmful commands.

  3. What to watch

    Open source models now account for more than 70% of tokens processed through Vercel's gateway and nearly 25% of total spend.

WHO IT HITSAI platform teams and developers choosing which models to route through API gateways are the first to feel this shift, since Jev's cheaper, faster classification approach could change how they allocate spend between frontier and open source models.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Context & Analysis

Vercel's data offers the first hard numbers behind the buzz that followed Jev's limited release on September 15. Published through Vercel's AI gateway, a unified API service similar to OpenRouter, the figures show 35% of paid teams using the model within two weeks — close to double the reach of Sonnet 5.5 and, by Vercel's measure, the fastest two-week adoption any model has recorded. Vercel's own decision to swap out GPT-6 Luna for Jev in classifying harmful commands suggests the appeal is practical rather than purely experimental.

The report also captures a broader split inside the gateway's traffic. Open source models now handle more than 70% of tokens processed and nearly 25% of total spend, while spending on frontier labs has grown tenfold even as it represents a smaller share of overall spend. That combination points to a market where the two categories may be complementary rather than competitive, though Vercel's gateway view is limited and may undercount labs that offer their own APIs.

FAQ
What is Jev designed to do?
Jev is a lightweight tool for choosing between a few predetermined outputs, which lets it run cheaper and faster than alternatives like Sonnet. That makes it suited to classifying large quantities of data quickly.
How was Jev released?
Jev was published in limited release on September 15 by TypeSafe AI.
How does Vercel use Jev itself?
Vercel was an early adopter, using Jev to replace GPT-6 Luna for classifying harmful commands.

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleReflection AI launches Beam, a 501B open-weight coding model