返回
原文
返回
AI 评分 46/100在 X 查看原推
导出 Markdown

CLM-8B 发布：比 Jev 快 9 倍的 System One 模型

elvis · @omarsar0 · X·2026-09-24 23:06·23分钟前
AI 导读

Contrastive Language Model（CLM）发布，这是一个用对比学习目标训练的 System One 模型，CLM-8B 推理速度最高比 Jev 快 9 倍，在 computer-use、游戏和工具调用任务上性能相当。

elvis@omarsar0
46AI 编辑部评分，满分 100

CLM-8B 发布：比 Jev 快 9 倍的 System One 模型

2026-09-24 23:06· 23分钟前
AI 导读

Contrastive Language Model（CLM）发布，这是一个用对比学习目标训练的 System One 模型，CLM-8B 推理速度最高比 Jev 快 9 倍，在 computer-use、游戏和工具调用任务上性能相当。

Pay attention to this new wave of System One models if you are building custom harnesses.

First Jev. Now, Contrastive Language Model (CLM).

CLM is 9x faster than Jev.

CLM seems to be a better verifier than Jev, particularly at long-horizon tasks.

How do Jev and CLM differ?

CLM is contrastive, and Jev is trained with Reinforcement Learning for Calibrated Decisions (RLCD).

Jev receives a situation plus predefined questions, and returns typed decisions with probabilities. CLM embeds the situation and candidate actions, compares their similarity, then ranks or selects the best match.

The point is that there are several ways to attack this problem, which is exciting.

You can see my recent guide on combining System One and System Two models for building custom harnesses. https://academy.dair.ai/resources/jev-decisions-in-a-pi-sdk-harness

Jacky KwokIntroducing Contrastive Language Model (CLM): an ultra-fast System One Model trained with a contrastive learning objective that connects states and actions. CLM...

来源：elvis· x.com

导出 Markdown