Latent Space·· 15 d ago
[AINews] Jev: a “System One Model” that only decides/classifies/routes/scores — >100x faster, >200x cheaper than small frontier LLMs
[AINews] Jev: a “System One Model” that only decides/classifies/routes/scores — >100x faster, >200x cheaper than small frontier LLMs
AI summary
TypeSafe released Jev, a non-autoregressive model trained with RLCD solely for decisions, classification, routing and scoring. Official claims put it at 20–200 times faster and 40–400 times cheaper than small frontier LLMs, with no output token charges.
Selection record
Threshold 76Media and individualsFirst 72Second 62
Not admittedSum of both 134 < twice the threshold 152
- Source tier
- Media and individuals; this tier's threshold is 76
- Pre-filter
- passed：标题及正文详述AI决策模型与评测
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: Latent Space · latent.space