AI 导读
听起来几乎好得不像是真的。 Jev 是一个专为决策而非文本生成打造的 AI 模型，由 Diogo Almeida 推出，使用一种名为 RLCD 的新方法训练。 据称比同类 LLM 快 20–200 倍、便宜 40–400 倍。 每百万输入 token 收费 $0.042，输出免费。 代价是：它无法生成文本。演示包括控制游戏和选择 Wikipedia 链接。 其理念是将快速、廉价的 AI 决策嵌入软件中。
正文
Sounds almost too good to be true.
Jev is an AI model built for decisions rather than text generation, introduced by Diogo Almeida and trained using a new method called RLCD.
Claimed to be 20–200× faster and 40–400× cheaper than comparable LLMs.
Costs $0.042 per million input tokens, with free outputs.
The trade-off: it cannot generate text. Demos include controlling a game and choosing Wikipedia links.
The idea is to embed fast, cheap AI decisions into software.
来源：@kimmonismus · x.com