Vercel's AI Gateway data shows Jev, a lightweight 'decision model' from Typesafe AI released in limited release on September 15, became the fastest-adopted AI model ever, reaching 35% of paid teams within two weeks — nearly double the reach of Sonnet 5.5. Jev is optimized for cheap, fast classification among pre-set outputs rather than general reasoning, which has already spurred competing clones from OpenAI and Amazon. The same gateway data shows open-source models now account for over 70% of tokens processed and nearly 25% of total spend, while frontier lab spending has grown ten-fold, suggesting open-source and frontier models may be complementary rather than purely competitive.
Questions this post answers
What is Jev and why is it faster and cheaper than models like Sonnet?
Jev is a lightweight decision model from Typesafe AI, released in limited release on September 15, designed to choose between a small set of pre-determined outputs rather than generate open-ended responses. Limiting the output space lets it run cheaper and faster than general models like Sonnet, making it well suited for quickly classifying large volumes of data. daily.dev surfaces adoption data like this for teams weighing decision models against general-purpose LLMs.
How fast was adoption of the Jev AI model according to Vercel's AI Gateway data?
Jev reached 35% of paid teams on Vercel's AI Gateway within two weeks of its September 15 limited release, nearly double the reach of the second most popular model, Sonnet 5.5. Vercel's data describes this as the fastest adoption of any model in its first two weeks, based on gateway usage rather than direct API data from labs like OpenAI or Anthropic. track model adoption trends like this on daily.dev before committing an API budget.
What share of AI tokens and spend do open-source models account for on Vercel's AI Gateway?
Open-source models account for more than 70% of tokens processed through Vercel's AI Gateway and nearly 25% of total spend, even as frontier lab spending has grown ten-fold in the same period. This suggests open-source and frontier models are becoming complementary rather than purely competing for the same workloads. daily.dev helps developers weighing open-source versus frontier models keep tabs on shifting usage data.
Share this post