Event ·Multi-source

Claude 发布 Fable 5.2 与 Mythos 5.2，多项能力超 Opus 5.5

Claude 同时发布 Fable 5.2 与 Mythos 5.2 两款新模型，面向编程、数学证明、生物研究和机器人控制等高难度技术任务。官方称两款模型在几乎所有评估能力上超过 Opus 5.5，部分任务的成本仅为原来的一半。升级覆盖编码、推理和视觉三条主线，重点不再是单一基准跑分，而是完成更长链条的复杂任务。对开发者而言，能力提升叠加成本下降，意味着长任务工作流（如大型代码库重构、自动化实验）的性价比发生变化，实际效果仍需以独立基准和真实项目验证为准。

Current read

Claude 一次放出 Fable 5.2 和 Mythos 5.2 两款模型，官方说几乎全面超过 Opus 5.5，部分任务成本还砍半。做长链条技术任务——大项目编码、数学证明、机器人控制——的团队可以留意这次能力覆盖面。

11 sources33° AI heatLast updated 2026/11/10 00:30:34

Evidence

11 sources

Primary sourceAI Will

2026/11/10 00:30:34View source

Related source 1Mike Krieger

2026/11/9 19:04:01View source

Related source 2rohanpaul_ai

2026/11/9 19:09:53View source

Related source 3小互

2026/11/10 00:50:04View source

Related source 4Claude

2026/11/9 18:58:01View source

Related source 5Cursor

2026/11/9 19:52:29View source

Related source 6Alex Albert

2026/11/9 19:09:10View source

Related source 7lmarena.ai

2026/11/9 19:15:48View source

Related source 8kimmonismus

2026/11/9 19:55:00View source

Related source 9claudedevs

2026/11/9 19:04:17View source

Related source 10Bindu Reddy

2026/11/9 19:08:08View 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.

Open the story