Jev: the model built to be called by code, not by humans

Key takeaways

1. 1

   Jev belongs to a class of models TypeSafe calls "system one," "machine native," or "large programmable": the consumer of the output is code, not a human.

2. 2

   Only three primitives — nule (derived from Bernoulli), score, and choice — mapping to an if, a sort/threshold, and a switch on an enum.

3. 3

   The model is optimized for "intelligence per dollar" (hence the name, a nod to Jevons' paradox), not for speed or leaderboard rankings.

4. 4

   Almeida calls himself "extremely anti public benchmarks": too easily gamed, they should give way to internal evaluations by developers on their own workflows.

5. 5

   No refusals in the API: "refusal is obviously a type error" — a random refusal in a dependency running in the background breaks the software.

6. 6

   Data is entirely synthetic, and the company refuses to train on user data, to avoid overfitting to the present.

7. 7

   Cited figures: over 1 trillion tokens per day, including overnight, 100,000 people on Discord, and almost no revenue before launch.

Chapters