Jev versus text-generating models
Jev does not generate free-form text. A side-by-side video compares parallel decision-making with token-by-token text generation.
Jev Harness: Reading and searching take 56.2% of tool turns, coding under 10%
Diogo Amogo's Jev Harness PDF proposes gating commands by trust, loading instructions by condition, and sharing one retrieval pass across read-only tasks, arguing the bottleneck was never coding.
Reading and searching eat 56.2% of tool turns. Writing code is under 10%. Diogo Amogo's Jev Harness PDF flips that: 1. gate commands by trust 2. load instructions by condition 3. share one retrieval pass across read-only tasks The bottleneck was never the coding.
· 0 likesOpen on X
More like this
Pricing, evaluations, and the name Jev
The launch thread explains code-and-AI workflow evaluations, the input price announced at launch, and free output tokens.
Jev Model: Decision-Making Probability Generation, 200× Faster and 400× Cheaper Than LLMs
This model is built for decision-making rather than text generation, outputting probabilities for each option in parallel instead of token-by-token responses. The author shares a simplified comparison of normal LLMs versus this Jev model.
