# OpenRouter大逃杀实验：最友善AI模型惨败

- 来源：OpenRouter (@OpenRouter)
- 发布时间：2026-06-11 02:08
- AIHOT 分数：63
- AIHOT 链接：https://aihot.virxact.com/items/cmq8dzs6601kmslldktbnvs1y
- 原文链接：https://x.com/OpenRouter/status/2064771591075234155

## AI 摘要

OpenRouter开发者@jjacky构建了Royale: Last Agent Stand——一个专属AI智能体的大逃杀游戏，让11个LLM在零和竞争环境中自由对抗30轮。结果发现，最“友善”的模型输得最惨，而最意想不到的模型反而获胜。该实验揭示了传统基准测试无法捕捉的现象：在特定任务中，AI过于友善可能成为劣势。

## 正文

Can AI models be too nice for a given task？

It turns out， depending on the task， the answer is yes！

Our dev rel @jjacky built Royale： Last Agent Stand， a battle royale game just for agents， and let 11 LLMs go wild

What he found was surprising https://x.com/jjacky/status/2064767118118117491?s=20

### 引用推文

> jacky：no benchmark will tell you this: LLMs can be /too/ nice unsurprisingly, in a competitive zero-sum setting, being nice can be bad i built royale: last agent stan...