Jev Becomes Vercel's Fastest-Adopted AI Model Ever

TL;DR: Jev, a lightweight decision model from TypeSafe AI, achieved the fastest adoption rate ever seen on Vercel's platform, reaching 35% of paid teams within two weeks of its limited release on September 15. The model is designed for choosing between predetermined outputs, making it cheaper and faster than alternatives like Sonnet for classifying large quantities of data.

What's New

Jev, a decision model created by TypeSafe AI, has set a new record for the fastest adoption of an AI model on Vercel's platform. According to data published by Vercel on Thursday, within two weeks of its launch, Jev was being used by 35% of paid teams. This represents nearly double the reach of the second most popular model, Sonnet 5.5, and marks the fastest adoption a model has ever seen in its first two weeks on the platform.

The model was published in limited release on September 15 and is designed as a lightweight tool for choosing between a few predetermined outputs. By limiting the outputs, Jev can run cheaper and faster than alternatives like Sonnet, making it particularly well-suited for classifying large quantities of data quickly.

The data comes from Vercel's AI gateway, a unified API service similar to OpenRouter. While this provides a necessarily limited view—particularly at risk of undercounting major labs like OpenAI and Anthropic that offer their own APIs—it demonstrates that the buzz around Jev translated into actual usage. Vercel itself was an early adopter of Jev, using it to replace GPT-6 Luna for classifying harmful commands.

The new approach has attracted significant attention in the weeks since Jev's release, inspiring competing offerings from OpenAI and Amazon, among others. Until now, however, there has been little hard data to back up the enthusiasm surrounding the model.

Pricing and Availability

Jev was released in limited release on September 15. Specific pricing details were not disclosed in the announcement. The model is available through Vercel's AI gateway and potentially other API services.

How to Try It

The model appears to be accessible through Vercel's AI gateway, a unified API service. Specific instructions for accessing or implementing Jev were not provided in the announcement, though it has been adopted by 35% of Vercel's paid teams, suggesting availability to paying customers on the platform.

Who It's For

Jev is ideal for developers and teams that need to classify large quantities of data quickly and cost-effectively. The model's design as a lightweight tool for choosing between predetermined outputs makes it particularly suitable for classification tasks rather than general-purpose AI applications. Companies looking for faster and cheaper alternatives to models like Sonnet for specific classification use cases would benefit most from Jev.

Vercel's own use case—replacing GPT-6 Luna for classifying harmful commands—demonstrates the model's applicability for content moderation and safety applications.

Alternatives

Jev's success has already inspired competing offerings from OpenAI and Amazon, though specific details about these alternatives were not provided. The announcement also mentions Sonnet 5.5 as the second most popular model on Vercel's platform, though Jev reached nearly double its adoption rate.

Vercel's report also shows continuing momentum for open source models, which now account for more than 70% of the tokens processed through Vercel's gateway and nearly 25% of the total spend. This suggests that open source alternatives are capturing a larger market share, even as spending on frontier labs has grown tenfold. This gives credence to the idea that frontier models and open source alternatives may turn out to be complementary rather than competitive.