跳到主内容
@wquguru
精选75Chubby♨️行业动态多源精选 ×7

Anthropic IPO估值2万亿美元,OpenAI成最大输家

A quick take on Anthropic’s IPO and how it ties in with OpenAI.

原文
发到 X

A quick take on Anthropic’s IPO and how it ties in with OpenAI.

简要谈谈 Anthropic 的 IPO 及其与 OpenAI 的关系。

Anthropic made history today, and I think it’s safe to say the IPO was a success across the board. Its fully diluted valuation reached about $2 trillion at the start of trading. Shares opened at $111.50, 34.3% above the $83 IPO price, and stayed above the IPO price in subsequent trading.

Anthropic 今天创造了历史,我认为可以说其 IPO 在各方面都取得了成功。其在交易开始时的完全稀释估值达到约 2 万亿美元。股票开盘价为 111.50 美元,比 83 美元的 IPO 价格高出 34.3%,并在后续交易中保持在 IPO 价格之上。

Anthropic got everything it could have hoped for from its IPO and expects $104.8 billion in net proceeds from the IPO and the concurrent private placement. That’s fresh capital it needs and plans to use, in particular, to keep leasing data centers. That makes SpaceXSI an indirect beneficiary. In other words, Anthropic’s IPO is a win for both Anthropic and SpaceX.

Anthropic 从其 IPO 中获得了所有期望的收益,并预计通过 IPO 和同步私募融资获得 1048 亿美元的净收益。这是它需要并计划使用的资金,特别是用于继续租赁数据中心。这使得 SpaceXSI 成为间接受益者。换句话说,Anthropic 的 IPO 对 Anthropic 和 SpaceX 来说都是一场胜利。

OpenAI is the loser here. And even if a lot of people don’t want to hear it, I think it needs to be said. OpenAI lost in this case regardless (!) of whether GPT-6.2 Astra and Sol really are that bad or have gotten significantly better since then. The reason is that Anthropic’s IPO achieved what OpenAI is aiming for. With its IPO, Anthropic has gained a head start in terms of financing and established a public valuation benchmark against which OpenAI will be measured should it pursue an IPO of its own. That’s a fact.

OpenAI 是这里的输家。即使很多人不想听,我也认为必须说出来。在这种情况下,OpenAI 输了,无论 GPT-6.2 Astra 和 Sol 是否真的那么糟糕,或者自那以后是否有显著改进。原因是 Anthropic 的 IPO 实现了 OpenAI 所追求的目标。通过其 IPO,Anthropic 在融资方面获得了先发优势,并确立了一个公开估值基准,如果 OpenAI 寻求自己的 IPO,将以此作为衡量标准。这是一个事实。

The stronger OpenAI had looked ahead of the IPO and the more convincingly it had shown investors that it was the real play, the less money would have gone into Anthropic and the more capital would be available to OpenAI next year. That’s what matters.

IPO 前 OpenAI 的表现越强,向投资者证明它是真正首选的力度越大,流入 Anthropic 的资金就越少,明年 OpenAI 可用的资本就越多。这才是关键所在。

The two are each other’s biggest competitors, and I can’t imagine a scenario in which OpenAI didn’t try, at least in some way, to undercut Anthropic’s valuation by putting out a release two days before Anthropic’s IPO. I think it’s pretty clear that it didn’t work. And (again) it makes no difference whether OpenAI’s model is good or bad.

两者互为最大的竞争对手,我无法想象 OpenAI 不在 Anthropic IPO 前两天发布消息以某种方式压低 Anthropic 估值的情景。我认为很明显这并没有奏效。而且(再次强调),OpenAI 的模型是好是坏并无区别。

What was disastrous about the release was the public perception. The impression was that it wasn’t a good release, and the market bore that out. As the IPO numbers show, it posed no threat to Anthropic.

这次发布的灾难性后果在于公众认知。给人的印象是这次发布并不好,市场也证实了这一点。正如 IPO 数据所示,它对 Anthropic 构不成威胁。

If anything, it went the other way. In my view, OpenAI’s latest release probably hurt OpenAI and strengthened Anthropic, because for a lot of people, Opus 5.5 and Fable 5.2 are still the best models even after the release. And yes, I know Astra leads in many areas, including science. That carries little weight with the market or in the public eye. That’s how this IPO should be put in context. And that’s how the numbers should be read.

事实上,情况恰恰相反。在我看来,OpenAI 的最新发布可能损害了 OpenAI 并加强了 Anthropic,因为对许多人来说,即使在发布之后,Opus 5.5 和 Fable 5.2 仍然是最好的模型。是的,我知道 Astra 在许多领域领先,包括科学领域。但这在市场或公众眼中分量很轻。这就是应该将此次 IPO 置于其中的背景。这也是解读这些数据的方式。

Big day for Anthropic, a lot of work for OpenAI. But the end is still open.

Anthropic迎来重要一天,OpenAI则面临大量工作。但结局仍未定。

更进一步:量化金融体系

看懂新闻只是起点——沿量化金融路径,把它变成能交付的工程能力

进入量化体系 →