# 用户实测多模型路由：效果远逊GPT 5.5

- 来源：Peter Steinberger 🦞 (@steipete)
- 发布时间：2026-06-22 15:31
- AIHOT 分数：39
- AIHOT 链接：https://aihot.virxact.com/items/cmqowdwfa04sbslx6f185qwnr
- 原文链接：https://x.com/steipete/status/2068960117253632160

## AI 摘要

@LLMJunky 实测某多模型路由服务，5小时使用额度在1个prompt内即告罄。在threejs任务（构建Rocket League副本）中，生成效果远差于GPT 5.5，需7-8次来回通过Codex修复才勉强可玩；GPT 5.5一次完成且无需后续调整，Fable同样表现出色。该路由性能不及Mythos，早期印象不佳。

## 正文

I was skeptical about the multi-model routing. Seems my hinch was right.

### 引用推文

> am.will：I tried this so you don't have to. I know this is going to absolutely shock you but no this does not match the performance of Mythos. A few early thoughts: 1. T...