Anthropic 的 Opus 5 主打 Token 效率，而非能力跃升

今天，Anthropic 推出了 Opus 5。这是该模型的最新更新版本；Opus 近期已成为编程、软件开发及其他任务中的热门选择。

尽管这对 Opus 而言是一次值得关注的提升，但它似乎并不是一次达到 Opus 4.5 级别的 Agent 编程性能突破。

Anthropic 制作的一张图表涵盖 Frontier-Bench、DeepSWE 等多项基准测试，显示 Opus 5 在编程任务中的表现大致与备受吹捧的 Anthropic Fable 模型持平，或略有领先。表面上看，它在几乎所有类型的任务中都击败了 Opus 4.8 以及 OpenAI 的竞品 GPT-5.6-Sol。

各项基准测试反映的是性能的渐进式提升，而不是一次剧烈飞跃；其主要卖点在于，模型能够以大约一半的成本，提供略低于 Fable 的能力。

还值得一提的是，Anthropic 特意没有为 Opus 5 提供网络安全任务上的前沿训练，因此它在这方面远远落后于 Fable 和 Mythos。Anthropic 声称，它较擅长发现网络安全漏洞；但由于模型训练中的相关决策，它在对这些漏洞进行利用方面“明显落后于 Mythos 5”。

因此，Opus 5 并不具备 Fable 的所有同类争议性保护措施，例如在发生事件时为便于审查而将数据保留 30 天的政策。

Today, Anthropic rolled out Opus 5, the newest update for the model that has recently become a popular choice for coding and other software development tasks, among other things.

While this is a noteworthy bump for Opus, it doesn’t seem to be an Opus 4.5-level breakthrough in agentic coding performance.

A chart (made by Anthropic) with various benchmarks like Frontier-Bench and DeepSWE shows Opus 5 performing at about the same level or slightly ahead of the much-ballyhooed capabilities of Anthropic’s Fable model for coding tasks. It ostensibly beats Opus 4.8 and OpenAI’s competing GPT-5.6-Sol in just about every kind of task.

The various benchmarks show an iterative increase in performance, but not a radical leap, while the main pitch is that it’s a model that offers something just shy of Fable at approximately half the cost.

It’s also worth mentioning that Anthropic specifically avoided giving Opus 5 cutting-edge training on cybersecurity tasks, so it lags way behind Fable and Mythos in that regard. Anthropic claims it is relatively good at finding cybersecurity vulnerabilities, but because of decisions made in training the model, it is “substantially behind Mythos 5 on the exploitation of those vulnerabilities.”

As such, Opus 5 doesn’t have all of the same controversial protections that Fable had, such as the policy of keeping data for review for 30 days in case of an incident.