# Claude Opus性能骤降，Grok稳居榜首

- 来源：DogeDesigner (@cb_doge)
- 发布时间：2026-04-12 23:39
- AIHOT 分数：26
- AIHOT 链接：https://aihot.virxact.com/items/cmnxjn67y0092sl9ogsqjxi8j
- 原文链接：https://x.com/cb_doge/status/2043353392853221811

## AI 摘要

Anthropic的Claude Opus正在下滑。

最新基准测试显示，其准确率在短短几天内从83.3%降至68.3%。

这在编码过程中的幻觉率出现了大幅飙升。

Grok 4.20仍保持第一的位置。未被超越。https://t.co/FA5nbKKeS0

## 正文

Anthropic's Claude Opus is FALLING.

Latest benchmarks show its accuracy dropped from 83.3% → 68.3% in just days.

That's a major spike in hallucinations during coding.

Grok 4.20 still holds the #1 spot. Undefeated.