# BDH-CQ以0.0007美元成本登ARC-AGI

- 来源：elvis (@omarsar0)
- 发布时间：2026-08-11 22:56
- AIHOT 分数：37
- AIHOT 链接：https://aihot.virxact.com/items/cmsosxt5l05lrrohd22pj1bxy
- 原文链接：https://x.com/omarsar0/status/2087191397111656738

## AI 摘要

这有点不可思议。

BDH-CQ 在 ARC-AGI 1 上得分 29.5%。还行。但每个任务仅花费 0.0007 美元。

它的实现方式尤其有趣。它在潜在空间中进行循环推理，而非使用 CoT。他们还验证了类似 Transformer 的扩展，参数量最高达 600B。

## 正文

This is kind of wild.

BDH-CQ scored 29.5% on ARC-AGI 1. Fair. But with just $.0007 per task.

How it's done is particularly interesting. It reasons recurrently in latent space rather than using CoT. They've also verified Transformer-like scaling up to 600B params.

### 引用推文

> Zuzanna Stamirowska：Pathway's BDH-CQ model redefines the Cost-Efficiency Frontier on ARC-AGI-1: $.0007 at 29.5%. This is made possible by in-context learning and latent reasoning. ...