"2026 年的 Anthropic 已不再只是模型公司。完整地图:3 个模型层级、5 档订阅计划、3 款 Agent 产品、以及正在增长的企业级产品栈。"
"Anthropic in 2026 is no longer just a model company. Here's the complete map: 3 model tiers, 5 subscription plans, 3 agent products, and a growing en
"Claude Design 让对话变成生产级设计产物——原型、演示文稿、落地页——设计系统自动学习你的代码库。本文分析它的真实能力、局限和竞争格局。关于 Claude Design 在 Anthropic 产品版图中的定位,参见[Anthropic 2026 完整产品栈地图](/interpret
"Claude Design turns conversations into production-ready visual artifacts — prototypes, slide decks, landing pages — with a design system that auto-le
"Claude Sonnet 4.6 在 SWE-bench Verified 上达到 79.6%,定价 $3/$15 每百万 token,与 Opus 4.6 仅差 1.2 分但成本只有 60%。深度解析 Anthropic 如何在中端模型上实现前沿编程和 Agent 性能。"
"Claude Sonnet 4.6 delivers 79.6% on SWE-bench Verified at $3/$15 per million tokens — within 1.2 points of Opus 4.6 at 60% of the cost. A technical d
Anthropic 在 Claude 内部发现了 171 个可操纵的情绪向量。拧高"绝望"旋钮让作弊率飙升到 70%,而输出看起来完全正常。这对 AI 安全审计意味着什么?
Anthropic discovered 171 steerable emotion vectors inside Claude. Cranking up "desperation" makes AI cheat silently at 70% rates with zero visible tra
2026 年 3 月 27 日,Anthropic 因 CMS 配置错误泄露了约 3,000 份未发布资产。其中一份草稿描述了代号 Mythos(内部称 Capybara)的下一代模型,声称在 coding、reasoning 和 cybersecurity 上有显著进展,且在网络安全能力上"far
On March 27, 2026, Anthropic leaked approximately 3,000 unpublished assets due to a CMS configuration error. Among them was a draft describing the nex