# A big problem with this is that we don't really have a clear description of what mythos capabilities…

- 来源：Nathan Lambert (@natolambert)
- 发布时间：2026-04-19 02:40
- AIHOT 链接：https://aihot.virxact.com/items/cmo4pieq400rqslthaxtti5u2
- 原文链接：https://x.com/natolambert/status/2045573304518590567

## AI 摘要

这里的一个大问题是，我们并没有清晰界定 mythos capabilities 到底是什么。

发布博客中的每个基准测试都有模型能达标，当然。

但要说有模型能直接替换到相同用例中且性能毫无下降？我对此表示怀疑。

## 正文

A big problem with this is that we don't really have a clear description of what mythos capabilities are.

A model on each of the benchmarks in the launch blog post， sure.

A model that you can swap right in for the same use-cases and notice no drop in perf？ Doubt it.

### 引用推文

> rohit：Dario seems to think China and open source will hit Mythos capabilities in 6-12 months