Event ·Multi-source

网友晒未发布模型基准截图,称能力接近 Anthropic Mythos 5.2 的两倍

X 用户 kimmonismus 发布推文,称一份未发布模型的基准成绩接近 Anthropic Mythos 5.2 的两倍,并附上基准截图,用 acceleration 和 exponentials 两个词形容进展速度。该模型尚未正式发布,推文未给出模型名称、参数规模或具体基准分数。 这份成绩来自一张外链截图,无法独立核验测试环境、提示词设置和是否为官方数据。Mythos 5.2 本身也未出现在 Anthropic 官网公开模型列表中,推文中的对比基准线存疑。 未发布模型的截图对比在 X 上流传频繁,常见于测试者提前泄露或夸大宣传。开发者在据此选型前,应等待官方模型卡和可复现的评测数据。

Current read

X 上有人放出一个未发布模型的基准截图,说接近 Anthropic Mythos 5.2 的两倍,但模型名和分数都没公开。截图没法核验,先当八卦看,选型还是等官方数据。

11 sources27° AI heatLast updated 2026/11/12 18:21:25

Evidence

11 sources
Primary sourcekimmonismus
View source
Related source 2Rowan Cheung
View source
Related source 3Simon Willison’s Weblog
View source
Related source 4TestingCatalog
View source
Related source 6Gary Marcus
View source
Related source 7rohanpaul_ai
View source
Related source 8Cognition
View source
Related source 9techcrunch
View source
Related source 10@koltregaskes
View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.