METAMeta
15看多0看空27提及
Serenity @aleabitoreddit
专注 AI 光通信与半导体供应链研究，擅长从产业链与公司基本面寻找高赔率机会。长期跟踪 CPO 与光模块产业链，持续输出 SIVE、AAOI、LITE、POET 等标的的深度观点。
结论速览
META 整体偏多（看多 15 · 看空 0）
Facebook的论坛不太可能有效地与Reddit等成熟平台争夺用户。
最近发言：2026/08/08 18:27 · 无信心分
8-10 分观点使用大徽标▲ 看多▼ 看空● 提及
历史观点（42 条，最新在上）
- English · 原文Today I'm writing a weekend guide on how to do DD when shorting $NBIS: First, you look at hyperscaler earnings for AI cloud read through: > $GOOGL: reports record AI cloud demand + backlog + margin increases from earnings > $AMZN: reports record AI Cloud demand + backlog + margin increases from earnings > $META: reports higher than expected prices for available capacity from earnings. Now, time to look at Nebius: -> $NBIS: Growing hundreds of percent to $7-9B ARR by Q4. Growing margins, and guided 4GW+ contracted power. -> Sees Uber/Waymo splitting, putting more focus on Avride -> Sees Clickhouse growing rapidly every quarter. Okay looks bad! But next, you need a hedge? -> Wow! A $NIKE brand executive, after the stock dropped 75% over the past 5 years, went to $LULU to save that brand next? Lululemon seems good. Conclusion: Short Nebius and go long on $LULU中文 · 翻译今天我来写一篇周末指南，教你做空 $NBIS 时怎么做尽职调查（DD）： 首先，你要看看那些超大规模云服务商（hyperscaler）的财报，从中解读AI云业务的表现： > $GOOGL：财报显示AI云需求创纪录、积压订单量大增，利润率也提高了。 > $AMZN：财报显示AI云需求创纪录、积压订单量大增，利润率也提高了。 > $META：财报显示，可用产能的价格超出了预期。 好了，现在来看看 Nebius： -> $NBIS：到第四季度，年化经常性收入（ARR）将增长几百个百分点，达到 70-90 亿美元。利润率也在增长，并且指引了 4GW+ 的合同电力。 -> 看到 Uber/Waymo 正在分拆，更多精力放在 Avride 上。 -> 看到 Clickhouse 每个季度都在快速增长。 好了，看起来不妙！但接下来，你需要对冲吗？ -> 哇！一个 $NIKE 的品牌高管，在 $NIKE 股票过去五年跌了 75% 之后，接下来竟然去了 $LULU 拯救那个品牌？Lululemon 看起来不错。 结论：做空 Nebius，做多 $LULU。
- English · 原文As a earnings recap: $AMZN, $META, $GOOGL, and $MSFT guided a 2026 combined capex forecast to ~$720-$745 Billion. Up from $695-$725B Billion previously. Amazon: $220B Google: $195B-$205B Meta: $130B-$145B Microsoft: $175B We've already seen significant deleveraging and retail/institutional margin liquidations (maybe it continues for more time, who knows). But medium-long term, I'm not quite sure how anyone can be bearish the upstream semis or neoclouds. Given each hyperscaler has flagged either compute shortages, rising cloud demand + pricing power, or increased spending for chips/networking. My "bottleneck" thesis with many of these upstream semi supply chains is that when trillions in capital flows into things from InP substrates or memory (which were both treated as cheap commodities) or even energy. Lot of these current AI names that were treated as useless before in telecom cycles or even toilet sellers. Gets rerated when their inflection period hits and capex flows into their balance sheets. As seen with $NVDA GPUs past few years, memory this year, CPUs/MLCCs next few years, CPO in 2027, Glass Substrates in 2027, 800V in 2027, and so on.中文 · 翻译财报总结一下：$AMZN、$META、$GOOGL 和 $MSFT 预计，到2026年，它们的合计资本开支将达到约7200亿至7450亿美元。 比之前预期的6950亿到7250亿美元有所提高。 亚马逊：2200亿美元 谷歌：1
- English · 原文Just a TLDR of this week: Media: Hyperscalers overbuilt and are selling excess compute. -> $GOOGL, $MSFT, $META, and $AMZN: Demand for compute far exceeds capacity. Capex go brrr. Institutions: Emergency rate hike or 3x rate hikes, it's all over. -> Kevin Warsh: No rate Hike. Retail/Media: AI is crashing because it's a bubble. -> Citadel looking to buy AI leveraged hedge funds positions: 👀 Jim Cramer: Sell all your leveraged DC stocks. -> Koreans: SK Hynix +30%, Samsung +26.81%中文 · 翻译本周总结一下： 媒体说法：超大规模云服务商建多了，正在甩卖多余算力。 → $GOOGL, $MSFT, $META, 和 $AMZN：对算力的需求远超我们现有产能。资本开支一路猛增。 机构说法：紧急加息或加息三次，一切都完了。 → Kevin Warsh：不会加息。 散户/媒体：AI要崩盘了，因为这就是个泡沫。 → Citadel 正在考虑买入那些加了AI杠杆的对冲基金仓位：👀 Jim Cramer：卖掉所有你加了杠杆的DC股票。 → 韩国人：SK Hynix +30%，Samsung +26.81%
- English · 原文Just some takeaways from $META | $MSFT earnings calls: Microsoft: - Expected to be FCF positive in 2027, despite the increase in capex (extremely positive for AI buildout that it's funded by operating income) - "Free cash flow was $19.6 billion, reflecting higher capital expenditures" - Quartely capex was $41B, roughly 2/3rds were "short lived assets, primarily CPUs and GPUs" - Expects capex spend will be over $50 billion for next quarter - Capex Guidance at ~$175 billion and 2027 capex roughly the same. Spending plans unchanged and in line. - "Extending the estimated useful life of our data centers from 15 to 25 years" - "We will be among the first cloud providers to deploy next generation rack-scale AI infrastructure based on $AMD Helios and $NVDA Vera Rubin" - "Customer demand continues to exceed available capacity" Meta: - Capex $130-$145 billion (narrowed range), from $125B-$145B. - Meta is receiving offers at a "significant premium" to what they paid for it (compute scarcity, positive for neoclouds like $IREN / $NBIS ) - Expects significant portion of compute (like the 1 GW DC in El Paso) to develop internal models. - Meta has multiple ROI-positive uses for additional compute across its core business (internally, not Meta Compute) - "Finally, we believe that overall industry capacity is going to remain tight for the foreseeable future" - "The industry has under-built historically for the wave of AI adoption, making existing capacity, including our own, extremely valuable" - Susan Li TLDR: - $MSFT and $GOOGL largely sustaining AI capex buildout while remaining FCF positive or through operating incomes. - $META flags available compute materially below demand at least through 2027. And $MSFT also flags compute demand far exceeds supply. - All three hyperscaler capex largely in line with Google hiking capex figures. AI selloff seems extremely overblown now, hyperscalers continuing capex in line (with Microsoft being FCF positive) or even hiked with $GOOGL. Compute scarcity is visible throughout every single hyperscaler ER.中文 · 翻译简单总结一下 $META 和 $MSFT 财报电话会的几点收获： 微软： - 预计到2027年自由现金流（FCF）转正，尽管资本支出（capex）在增加（这对于AI基础设施建设是天大的利好，因为它的资金来源是运营收入） - “自由现金流为19
- English · 原文$GOOGL now has 950M monthly Gemini users and processes 22B API tokens/min. Compared to 750M back in February. Absurd growth adding 200 million more users in a few months. Kinda explains why they ran out of compute, then cut allocations to $META and others. Very unlikely AI capex will slow down given.中文 · 翻译$GOOGL 现在每月有 9.5 亿 Gemini 用户，每分钟处理 220 亿 API token。 相比之下，2 月份是 7.5 亿。 几个月内增加了 2 亿用户，这个增长太惊人了。这也解释了为什么他们算力不够用，然后砍了分给 $META 和其他公司的算力。 考虑到这些，AI 资本开支不太可能放缓。
- English · 原文$META in talks to lease compute to Anthropic in a $10B dollar deal. Seems like they saw how profitable $SPCX $45B compute deal was. But just goes to validate Neocloud business models like $NBIS, $IREN, and co if hyperscalers are copying their homework. Source: NYT https://t.co/3Mu3UXQtFT中文 · 翻译$META 据说正在谈判，想把算力租给 Anthropic，这笔交易估计值 100 亿美元。 看来他们是看到 $SPCX 那笔 450 亿美元的算力租赁有多赚钱了。 不过这正好也验证了像 $NBIS、$IREN 还有其他这类新云业务模式的价值，毕竟连超大规模云服务商都在“抄作业”了。 来源：纽约时报 https://t.co/3Mu3UXQtFT
- English · 原文@SVTrivo No leaks yet on what companies $AAPL is buying. Regardless, my guess is that they're trying to become more self-reliant after what happened with $GOOGL + $META and OpenAI. Which would signal AI capex hikes, which markets definitely aren't expecting from someone like Apple.中文 · 翻译@SVTrivo 苹果 $AAPL 要收购哪些公司，目前还没啥风声。 不过，我猜他们可能是经历过 $GOOGL + $META 和 OpenAI 的事情后，想变得更自给自足。 这可能预示着苹果会增加 AI 资本开支，而市场肯定没想到苹果会这么做。
- English · 原文@michaelsikand DRAM 20%+ hike for next quarter, $SNDK LTAs with $META, $MU 16+ LTAs. Market: proceeds to sell off memory. 800G transceiver revision sharply upward? Lasers completely sold out into early 2029? Market: sells off photonics and laser companies.中文 · 翻译@michaelsikand DRAM 下季度涨价 20% 以上，$SNDK 与 $META 签订了长期供货协议，$MU 也有 16+ 份长期协议。 市场：接着就开始抛售内存股。 800G 光模块的出货预期大幅上调？激光器件的订单已经排到 2029 年初，完全售罄了？ 市场：抛售光子学和激光公司的股票。
- English · 原文$AAPL looks to acquire AI chip companies for running AI (Source: The Information) Right now, $MSFT, $META, Amazon, Google are carrying AI capex spend. But what if Apple joined the others after M&A? A possible scenario is that they revise capex largely upward for their own AI buildout. Since they probably witnessed Google cutting off Meta from compute constraints... or what happens when you partner with OpenAI for LLMs. Then learned how important it is to have your own infrastructure. This scenario would be quite bullish thematically from optical networking to foundries and something markets would not expect? We'll see what happens.中文 · 翻译$AAPL 打算收购 AI 芯片公司来运行人工智能 (来源：The Information) 目前，$MSFT、$META、亚马逊、谷歌正在承担大量的 AI 资本支出。 但如果苹果在并购之后也加入其中，会怎么样呢？ 一种可能的情况是，他们会大幅上调资本支出，用于建设自己的 AI。 因为他们可能目睹了谷歌因为算力限制而切断了 Meta 的资源……或者也看到了与 OpenAI 合作开发大模型时会发生什么。 然后就明白了拥有自己基础设施的重要性。 这种情景对相关主题投资会是重大利好，从光网络到晶圆代工都会受益，而且这可能是市场没有预料到的？我们拭目以待。
- English · 原文I don't see any fundamentally wrong. There's probably going to be large corrections from time to time flush out margin/leverage before things move higher. And this month seems like that time of year? $POET confirmed your big optical giants like $LITE, $COHR are completely sold out for the next 2 years, and likely into 2029 for photonics. Innolight confirmed 800g transceiver upward revisions 3 days ago, so that should be positive for $AAOI and the others. Samsung became the most profitable company in the world, and continues to project DRAM hikes for future quarters. $MU signed 16+ LTAs showing memory demand is structural... $META + hyperscaler capex plans are on the higher end of projections. I wouldn't conflate short term price movements with longer term trends. And as seen with $AEHR, recoveries tend to be extremely fast (eg. 1M of corrections wiped out overnight).中文 · 翻译我没看到任何根本性的问题。市场很可能会时不时出现大幅回调，用来清洗保证金/杠杆，然后才会继续上涨。 这个月感觉就像一年中这样的时刻？ $POET 证实了那些大型光学巨头，比如 $LITE、$COHR，未来两年订单都排满了，光子学领域可能要到 2029 年。 Innolight 证实了 3 天前上调了 800G 收发器预期，所以这对 $AAOI 和其他公司来说应该是利好。 三星成为了全球最赚钱的公司，并继续预测未来几个季度 DRAM 价格上涨。$MU 签署了 16+ 份长期协议（LTA），这表明内存需求是结构性的... $META 加上超大规模数据中心资本支出计划都处于预测的较高区间。 我不会把短期价格波动和长期趋势混为一谈。就像 $AEHR 的情况那样，市场恢复通常会非常快（比如，一个月的回调可能一夜之间就被抹平）。
- English · 原文$NBIS signs $1B+ compute agreement with Reflection AI, for GB300 access through 2029. Reflection also signed a multi-billion dollar agreement with $SPCX earlier. Interesting to say the least, seeing Nebius drop -5% off the news today. Also... counterparty to get this done kinda reminds me of OpenAI, where they might not have the funds to actually execute on these LTAs yet compared to $META or $MSFT. But generally positive long term developments, customer diversification was one of the core strengths of Nebius.中文 · 翻译$NBIS 和 Reflection AI 签署了一份10亿多美元的算力协议，获得到2029年的GB300算力。 Reflection 之前也和 $SPCX 签署了一份数十亿美元的协议。 至少可以说很有意思的是，今天 Nebius 竟然因为这消息跌了5%。 另外…这个交易对手方能达成这笔交易，有点让我想起 OpenAI，就是他们可能还没像 $META 或 $MSFT 那样，有足够的资金来真正执行这些长期协议。 但从长期来看，这总体上是积极的进展，毕竟客户多元化一直是 Nebius 的核心优势之一。
- English · 原文Nothing like an internal $META memo getting published. Showing AI ramping as usual: - LTAs signed with Samsung and $SNDK for memory - LTAs signed with Sumitomo Electric for fiber optics - Expected to deploy 7GW compute infra this year, and doubling in 2027. - as much as $145B capex spend this year “It plans to launch a chip about every six months through 2027”中文 · 翻译没有什么能比 $META 内部备忘录被公开发布更刺激的了。 从中可以看到 AI 业务一如既往地在加速： - 与三星和 $SNDK 签订了长期存储器供货协议 (LTA) - 与住友电工签订了长期光纤供货协议 (LTA) - 预计今年部署 7GW 计算基础设施，到 2027 年翻一番 - 今年资本支出高达 1450 亿美元 “它计划在 2027 年之前，每六个月推出一款芯片”
- English · 原文$META to build a new $10B DC in Canada to expand AI capacity. So much for the media framing of “Meta Compute”as overbuilding and cutting capex. https://t.co/0QgSls0s9o中文 · 翻译$META 将在加拿大新建一个 100 亿美元的数据中心，以扩大 AI 容量。 所以媒体之前把“Meta 计算能力”描述成过度建设、削减资本开支，这种说法也就不攻自破了。https://t.co/0QgLls0s9o
- English · 原文Just Bloomberg and $META doing damage control after crashing the market with Meta Compute framing: Spokesperson: "Meta is still hungry for even more computing power. It is still moving forward with plans for expensive new data centers and recently inked major computing deals with $CRVW, Google, $ORCL, and others." Just dropped that in with the Meta Muse announcement, and evenn threw in the "expensive" framing with DCs to signal capex. But little late given we're likely seeing a lot of margin liquidation cascades and heavy losses from media framing earlier.中文 · 翻译彭博社和 $META 只是在用“Meta Compute”的说法搞砸市场之后，做损害控制： 发言人：“Meta 对更多的算力依然渴望。 它仍在推进昂贵的新数据中心的建设计划，并最近与 $CRVW、Google、$ORCL 等公司签署了重要的计算交易。” 只是在 Meta Muse 发布会上顺便提了这一点，甚至还特意提到了数据中心“昂贵”来暗示资本支出。 但考虑到我们很可能正在看到大量的保证金清算潮，以及之前媒体报道造成的巨大损失，这有点晚了。
- English · 原文I personally think so. It feels like markets started selling off after that misleading Bloomberg $META compute article, and a couple of SemiAnalysis bear posts on $NVDA supply chains. Since algos + headline interpretation of “Meta excess compute” was cutting capex + dropping out of AI race. Combine that with excess leverage and here we are. Probably just takes an earnings call and few Trump tweets to get back to normal… But after some names already dropped 40-60% feels like it should bottom soon. Reversals the other way are pretty violent too.中文 · 翻译我个人觉得是这样。感觉市场开始抛售，是在那篇误导性的彭博社 $META 算力文章出来之后，还有SemiAnalysis发的几篇看空 $NVDA 供应链的文章。 因为算法交易加上对“Meta算力过剩”这种标题的解读，就变成了Meta要削减资本支出，退出AI竞赛。 再加上过度的杠杆，就到了现在这个地步。也许只需要一次财报电话会议和几条特朗普的推文，就能恢复正常了…… 但一些股票已经跌了40-60%了，感觉应该很快就要见底了。 反过来说，反弹起来也通常很猛。
- English · 原文One of the dumbest thematic selloffs I’ve seen to to date off: - $META compute news that’s not new - CPO delay report 1, that got refuted by $NVDA - CPO delay report 2, that got refuted by $NVDA High confidence, institutions will end up going long on the same names they’re bearposting after retail capitulates.中文 · 翻译这是我迄今为止见过的最愚蠢的主题性抛售之一，起因是： - $META 的计算相关新闻，但那根本不是什么新消息 - CPO 延迟报告 1，然后被 $NVDA 辟谣了 - CPO 延迟报告 2，结果又被 $NVDA 辟谣了 我很有信心，散户彻底投降后，机构最终会做多他们之前一直在发帖唱空的那些股票。
- English · 原文I’d like to give a special thank you to Bloomberg’s piece on $META compute. As well as the 2 back to back bear posts on CPO delays. But I believe thematically photonics, memory, energy, and physical AI will outperform in the long run. https://t.co/vghsG5H2w5中文 · 翻译我想特别感谢一下彭博社关于 $META 算力的那篇文章。 以及关于 CPO 延迟，连着发出来的两篇看空帖子。 但我相信，从主题投资的角度看，光子学、存储、能源和实体 AI 长期来看会表现更好。https://t.co/vghsG5H2w5
- English · 原文Looks like there's a high power cylindrical cell / BBU cell shortage (aka: bottleneck): - Samsung SDI supplies the cells to Simplo, which assembles them into BBUs for $META to $AMZN. - As demand from data centers has increased, production of Samsung SDI's cylindrical cells has also risen sharply. - Industry sources said Panasonic, Samsung are experiencing supply shortages for BBU cells. So main beneficiaries imo look like: - Samsung SDI (KRX: 006400) / Panasonic Energy as cleaneset winners - Murata (6981) / this keeps showing up everywhere with MLCC and others lol - LG Energy (KRX: 373220) / incoming There's not really much direct US players? But adjacent read through like $VRT, $ETN or $BWA / $ENS. Not quite in my domain emphasis cup of tea though. I wouldn't conflate this as all shortages having a massive TAM, but there could be opportunities...中文 · 翻译看来，大功率圆柱形电池 / BBU 电池好像短缺了 (也就是：瓶颈)： - 三星 SDI 向 Simplo 供应电池，后者将电池组装成 BBU，供货给 $META 和 $AMZN。 - 随着数据中心需求的增加，三星 SDI 圆柱形电池的产量也大幅上升。 - 业内人士称，
- English · 原文@fivepointscap Yep, $SE $MELI $JD $AMZN and the other e-commerce giants are the most obvious beneficiary of physical AI push that I can think of Given they can optimizing opex immediately with robotics. Names like $META are kinda less clear.中文 · 翻译@fivepointscap 对，$SE、$MELI、$JD、$AMZN 以及其他电商巨头，是我能想到的实体AI推进最明显的受益者。 因为他们可以用机器人即刻优化运营成本。像 $META 这样的公司就有点不那么明确了。
- English · 原文@softmaxedx I blame Semianalysis for crashing the entire photonics market for misleading CPO delay reporting that $NVDA refuted. Alongside global journal $META recent misleading reporting… But I think they realized everything was crashing along with it, so they needed a macro bull post.中文 · 翻译@softmaxedx 我要怪 Semianalysis，他们搞崩了整个光子学市场，就因为那份被 $NVDA 反驳的、误导性的 CPO 延迟报告。 再加上 $META 最近那些全球性的误导性报道… 但我觉得他们意识到所有东西都跟着在崩盘，所以他们需要发一个宏观看多的帖子。
- English · 原文@ontrialperiod when global and US markets crash off misinformation and out of context quotes from $META. I probably want to clear that up? That thing said, $AAOI, $SIVE, and $CCXI do make up large concentration in my portfolio.中文 · 翻译@ontrialperiod 当全球和美国市场因为 $META 的错误信息和断章取义的言论而暴跌时。我大概想澄清一下？ 话虽如此，$AAOI、$SIVE 和 $CCXI 确实在我投资组合中占了很大比重。
- English · 原文SemiAnalysis on $META “overcapacity” and market reactions with $NBIS and others: “We believe Meta’s datacenter and compute will accelerate”. “Capex in 2027 will be shockingly high”. Recent global crash, especially in the photonics sector was stupid… Off misleading narratives of Meta dropping out of AI race to sell excess compute… When in fact things are likely to accelerate from Meta catching up to GPT5.5. I’m personally expecting a sharp V recovery, especially with the names that crashed 50%+ from this narrative.中文 · 翻译SemiAnalysis 关于 $META “产能过剩” 和市场对 $NBIS 及其他公司的反应： “我们相信 Meta 的数据中心和计算能力将加速发展。” “2027 年的资本支出将高得惊人。” 最近全球市场的暴跌，尤其是在光子学领域，真是太蠢了……完全是基于 Meta 将退出 AI 竞赛、出售过剩算力的误导性叙述…… 而事实上，随着 Meta 追赶 GPT5.5，一切很可能都会加速。 我个人预计会有一个 V 形强劲反弹，特别是那些因为这个说法而暴跌了 50% 以上的公司。
- English · 原文Lot of misinformation recently about $META. Recently major journals have been stating “AI AGENT DEVELOPMENT OVER THE LAST FOUR MONTHS HASN'T 'ACCELERATED IN THE WAY WE EXPECTED” To push some narrative about Meta dropping out of the race or lowered capex: As Wang clarified: Zuckerberg was talking about the industry as a whole.中文 · 翻译最近关于 $META 有很多不实信息。 最近一些主流媒体一直在说： “过去四个月里 AI 代理的开发，并没有像我们预期的那样加速。” 他们这么说，是为了鼓吹一种说法，就是 Meta 要退出竞争，或者会降低资本支出： 但正如王所澄清的：扎克伯格当时说的是整个行业的情况。
- English · 原文@navborgen But market narratives were saying Meta would drop out of the AI race, stop capex, and sell excess compute!! Can’t believe that caused everything to crash.中文 · 翻译@navborgen 但当时市场都在说 Meta 会退出 AI 竞赛，停止资本支出，还会出售多余的算力呢！！ 真不敢相信那竟然让所有东西都崩盘了。
- English · 原文$META: upcoming AI model “Watermelon” has caught up to OpenAI’s GPT 5.5. Wang noted it uses an "order of magnitude" more compute than its predecessor “Avocado”. Who said they were out of the AI race again? https://t.co/X2sOgMRTuw中文 · 翻译$META：他们即将推出的 AI 模型“Watermelon”已经赶上了 OpenAI 的 GPT 5.5。 Wang 指出，它使用的算力比其前身“Avocado”高出“一个数量级”。 谁又说他们退出 AI 竞赛了？https://t.co/X2sOgMRTuw
- English · 原文Always a fun time seeing semi markets crash from Bloomberg report framing once again… UBS analyst on $META report: “this is not new news”. Regardless: $AEHR down -18.3% $AAOI down -17.13% $SIVEF down -15.4% $SNDK down -14.8% $TER down -13.8% $GLW down -11.4% $MRVL down -11.2% $LITE down -10.2% $NBIS down -7.8% Getting PTSD from Bloombert’s previous misleading report on AMD/Nvidia global export controls earlier this year that crashed semi stocks. FYI: Meta doesn’t randomly buy $48B+ worth of neocoud contracts if they overbuilt capacity and can cut capex… Markets are stupid sometimes.中文 · 翻译每次看到半导体市场因为彭博社的报道措辞再次暴跌，都觉得挺有意思的… 瑞银分析师评论 $META 财报说：“这不是什么新消息。” 但不管怎么说： $AEHR 下跌 -18.3% $AAOI 下跌 -17.13% $SIVEF 下跌 -15.4% $SNDK 下跌 -14.8% $TER 下跌 -13.8% $GLW 下跌 -11.4% $MRVL 下跌 -11.2% $LITE 下跌 -10.2% $NBIS 下跌 -7.8% 看到这个，我都想起彭博社今年早些时候那篇关于 AMD/Nvidia 全球出口管制的误导性报道，当时也搞得半导体股票大跌，简直要犯 PTSD 了。 顺便提一下：如果 Meta 真有过剩产能并且能削减资本支出，他们就不会随便买下价值超过 480 亿美元的 Neocoud 合同了… 市场有时候就是这么傻。
- English · 原文Photonics is backed by actual revenue numbers and it's an architectural shift championed by $NVDA. Quantum barely has any revenue. $LITE is completely sold out for the next 2 years (per $POET AGM) likely starting into 2029. Lumentum is so strained that they buy CW lasers off competitors (earnings transcript) $COHR is bottlenecked, so they buy EML off Lumentum. Then, $AAOI is coming in with Made-in-America independent CW capacity, are projecting $1.4B/quarterly revenue ending H1 2027 of a stupid $9.3B MC today. So all the CW capacity from independent players who have it now like $AAOI or $SIVE are likely to become scarce resources. Many other hyperscalers have already started LTA discussions (per Trendforce). And players like $AMD are currently talking with players such as $AAOI (Rosenblatt channel checks). Thematically, next 2 years is 9x TAM to US$154B per GS reports, especially with 16x/45x dollar content increase in scale out/scale up. Then there's the overall thematic AI drop from $META, which is widely misunderstood because people conflate what "excess capacity" means. And as UBS mentioned, Meta planning a cloud offering is NOT NEW NEWS. Bloomberg just has a tendency to publish information that causes doom drops across the semi sector like Nvidia export controls a few months back. But I'm familiar with what I'm holding so I'm confident in these numbers playing out. Especially when all the major players are sold out, the fundamentals catch up eventually.中文 · 翻译光子技术是有实实在在的营收数据支撑的，而且这是 $NVDA 大力推动的架构转型。 量子技术则几乎没什么营收。 根据 $POET 股东大会的消息，$LITE 未来两年（可能从2029年开始）的产品已经完全售罄。Lumentum 的产能非常紧张，甚至要从竞争对手那里购买连续波激光器（财报电话会议记录里提到了）。 $COHR 则卡在瓶颈上，所以他们向 Lumentum 购买 EML。 接着，$AAOI 正带着“美国制造”的独立连续波产能入局，他们预计到2027年上半年，季度营收将达到14亿美元，要知道它今天市值才93亿美元，简直太低了。 所以，现在像 $AAOI 或 $SIVE 这样拥有独立连续波产能的厂商，他们的产能很可能会变成稀缺资源。 很多其他超大规模数据中心厂商已经开始谈长期供货协议（来自 Trendforce 的消息）。而像 $AMD 这样的公司，目前正在和 $AAOI 等厂商洽谈（Rosenblatt 的渠道调研显示）。 从主题上看，根据 GS（高盛）的报告，未来两年市场总规模（TAM）将增长9倍，达到1540亿美元，特别是在横向扩展/纵向扩展方面，每美元内容投入将分别增加16倍/45倍。 然后还有来自 $META 的关于 AI 的整体主题性下跌，这被很多人误解了，因为大家混淆了“过剩产能”的真正含义。而且正如 UBS（瑞银）提到的，Meta 计划提供云服务并不是什么新鲜事。 彭博社就是有这种倾向，喜欢发布导致半导体行业全面暴跌的消息，就像几个月前关于英伟达出口管制的那次。 但我很清楚自己持有的是什么，所以我对这些数据会实现很有信心。 尤其是在所有主要玩家都产能售罄的情况下，基本面终究会追上来的。
- English · 原文@mkfilko Morgan Stanley for example is expecting an upward revision on AI capex from the $META announcement too… So selloff on all AI names seems like a bit much. https://t.co/UTyz9a7KrM中文 · 翻译@mkfilko 比如摩根士丹利也预期，随着 $META 的公告，AI 资本支出会向上修正… 所以所有AI相关股票都出现抛售，感觉有点过头了。https://t.co/UTyz9a7KrM
- English · 原文Wells Fargo: $META intent to sell excess compute is a positive signal around underlying demand and unit economics of AI. “Despite this shift, we don’t expect a pullback in Meta’s capex or that overall compute needs are lower” Regarding Neoclouds: WF thinks it validated the massive AI infra opportunity as well as acquisition opportunities. Despite any potential competition for Neoclouds. I’m inclined to agree with Wells Fargo here and say markets completely misunderstood Meta’s excess compute comment.中文 · 翻译Wells Fargo：$META 打算出售过剩算力，这对人工智能的底层需求和单位经济效益来说，是个积极信号。 “尽管有这种转变，我们不认为 Meta 会削减资本支出，也不认为他们整体的算力需求会降低。” 关于 Neoclouds：Wells Fargo 认为它验证了庞大的人工智能基础设施机遇，以及潜在的收购机会。即使 Neoclouds 可能会带来竞争。 我个人倾向于同意 Wells Fargo 的观点，认为市场完全误解了 Meta 关于过剩算力的评论。
- English · 原文There’s a lot of disinformation going around about $META “cutting capex” because they “overbuilt”. This is an “if” they have excess capacity. And it looks like the opposite right now: Hyperscalers like $GOOGL are so compute constrained that they had to cut allocations to Meta back in March. Since Meta was using too much for internal projects. Meta was immediately constrained so it looks like they were forced to immediately sign massive $48B+ contracts with Neoclouds like $CRWV and $NBIS. Meta is selling excess capacity if there’s any, especially since their large contracts are take or pay from the Neoclouds. If anything, I’m expecting their guided capex to go up as they build out more independent capacity.中文 · 翻译关于 $META “削减资本支出”是因为他们“超建了”的谣言很多。 这有个前提，那就是他们真的有过剩产能。 但现在看起来，情况恰恰相反： 像 $GOOGL 这样的超大规模云厂商，计算资源紧张到三月份就不得不削减给 Meta 的分配。 因为 Meta 用了太多资源来做内部项目。 Meta 立即受到了限制，所以看起来他们被迫立马和像 $CRWV、$NBIS 这样的“新云”公司签下了超过 480 亿美元的巨额合同。 Meta 如果真有闲置产能的话，那肯定是正在出售，尤其是考虑到他们和这些新云公司签的大合同都是照付不议的。 要我说的话，我反而预计他们指导的资本支出会增加，因为他们要建设更多自己的独立产能。
- English · 原文Finally, $META to build a Cloud Business to sell excess AI compute per Bloomberg. Thought Meta should go up against GCP, Azure, and AWS eventually to increase revenue streams. Meta up 6.65% premarket. https://t.co/UUolnwtYJY中文 · 翻译终于，$META 要建云计算业务了，据彭博社说，是想把过剩的 AI 计算能力卖出去。 我觉得 Meta 迟早要和 GCP、Azure、AWS 竞争的，这样能增加收入来源。 Meta 盘前涨了 6.65%。https://t.co/UUolnwtYJY
- English · 原文Guess we finally found why $META signed massive agreements with Neoclouds like $NBIS back in March... And why Gemini got super nerfed. $GOOGL reportedly restricted Meta's capacity in March 2026 because of compute restraints. Google's CEO said from last earnings computing power restrictions prevented Google Cloud from taking on more customer needs and made the department's backlog nearly double the previous quarter. This is probably positive for the AI DC capex buildout since hyperscalers capacity is way below what is needed, and especially so if they can't rely on one another.中文 · 翻译看来我们终于搞明白了，为啥 $META 早在三月份就跟 $NBIS 这类 Neocloud 签了巨额协议…… 也解释了为啥 Gemini 被砍得那么狠。 据报道，$GOOGL 在 2026 年 3 月因为算力限制，限制了 Meta 的容量。 Google 的 CEO 在上次财报电话会上说过，算力限制导致 Google Cloud 无法承接更多客户需求，使得该部门的积压订单环比几乎翻了一倍。 这对 AI 数据中心资本开支建设来说大概是利好，因为超大规模云厂商的产能远低于实际需求，尤其是如果它们彼此之间还指望不上对方的话。
- English · 原文Yes, choosing the right theme is extremely important. Even if $POET doesn't really do anything, it still gets brought up thematically due to $LITE, $COHR, and others. Even if $RDDT outperforms extremely hard, it still gets brought down from $META, $MSFT, $CRM, and other software basket names.中文 · 翻译没错，选对主题极其重要。 就算 $POET 本身没什么动作，它也会因为 $LITE、$COHR 和其他股票而被带到主题上来。 就算 $RDDT 表现极其强势，它还是会因为 $META、$MSFT、$CRM 以及其他软件篮子里的票而被拖下来。
- English · 原文Other way around, $NVDA bottlenecked the entire industry for EML capacity. And did the same with CW capacity ONCE AGAIN with $LITE, $COHR, and $MRVL (if they have LTA in place with Celestial) I said this a few months ago, we'd see this exact same playbook. But $AMD, $AMZN, $META, and others are just so stupidly slow that they let themselves get bottlenecked. Now there's only a few merchant players like $AAOI, $MTSI, and $SIVE that they all need to fight over.中文 · 翻译反过来看，是 $NVDA 卡住了整个行业在 EML 产能上的脖子。 而且它又干了一次同样的事情——在 CW 产能上，把 $LITE、$COHR 和 $MRVL 也给卡住了（前提是他们跟 Celestial 签了长期协议）。 我几个月前就说过，我们会看到完全一样的剧本。但 $AMD、$AMZN、$META 这些家伙动作慢得离谱，结果让自己也被卡在瓶颈里。 现在只剩少数几家商业玩家了，比如 $AAOI、$MTSI 和 $SIVE，大家都得争着抢。
- English · 原文At this point I can't tell anymore if markets from $META to $MSFT are correcting because of macro. Or just liquidity pull from $SPCX + index inclusion. And institutions frontrunning Nasdaq 100 and other rebalancing of SpaceX... Anyone know? https://t.co/sTLjsGq95V中文 · 翻译到现在我已经分不清， 从 $META 到 $MSFT 的市场是在因为宏观面回调， 还是只是 $SPCX 的资金流出 + 指数纳入造成的流动性抽离， 再加上机构抢跑纳斯达克 100 和 SpaceX 等其他再平衡操作…… 有人知道吗？ https://t.co/sTLjsGq95V
- English · 原文@GalV19634050 I don’t see any signs of capex slowing down, given both Meta and Google just did a raise, OpenAI raised a ton of + going public. As long as hyperscaler capex keeps ramping, many names above and my other CPO exposure names should heavily benefit.中文 · 翻译@GalV19634050 我看不出资本支出有任何放缓的迹象，毕竟 Meta 和 Google 刚融了一轮资，OpenAI 也融了一大笔钱，还要上市。 只要超大规模企业的资本支出还在加码，上面提到的那些标的，以及我其他 CPO 相关持仓，都应该会大幅受益。
- English · 原文@soulbiri1 I think only $IBIT / $XLU / $META / $CRCL are red since that mention. Maybe like 1-2 flat like $HOOD But 25 for 30 like $NBIS green, and many by triple digits is pretty solid if you do equal weighted.中文 · 翻译@soulbiri1 我觉得只有 $IBIT / $XLU / $META / $CRCL 自那之后是红的。 大概有 1-2 个像 $HOOD 这样是平的。 但 30 个里有 25 个像 $NBIS 是绿的，而且不少涨幅双位数，如果你做等权重配置的话，这成绩挺不错的。
- English · 原文Yeah… I think all your upstream semi supply chain companies are going much higher. Goldman now expects a combined $5.3 trillion of capex spending for the four largest hyperscalers $GOOGL / $META / MSFT / $AMZN from 2025 to 2030. Revised up from $4.5T from Q1 earnings. “Aggregate capex est. $7.6 trillion between 2026 and 2031.” And it flows upward to these tiny chokepoints like $SIVE for CPO lasers/ $SOI for Silicon Photonics substrates. Leaderdrive/Harmonic for Humanoids components. And so on… Ai names don’t move in a straight line up, but is just the beginning of the next Industrial Revolution as we move from R&D/compute buildout into commercialization from Agents -> Physical AI -> discovery.中文 · 翻译是啊……我觉得你上游所有半导体供应链的公司都会涨很多。 高盛现在预计，从2025到2030年，四大超大规模云厂商 $GOOGL / $META / MSFT / $AMZN 的资本开支总和将达到 5.3 万亿美元。 相比一季度财报时预测的 4.5 万亿，已经上调了。 “2026 到 2031 年，总资本开支估计为 7.6 万亿美元。” 这些资金会向上传导到那些小而关键的瓶颈环节，比如 $SIVE（CPO 激光器）、$SOI（硅光子学衬底），还有 Leaderdrive / Harmonic（人形机器人组件）。 等等等等…… AI 板块不会一路上涨， 但当我们从研发/算力建设阶段走向商业化——从 Agent 到物理 AI，再到科学发现——这仅仅是下一场工业革命的开端。
- English · 原文Just some random notes about $AVGO earnings transcript - Revenue target reiterated ($100B+ 2027, pretty sure markets wanted that to be raised this earning, hence the drop) Remember $NVDA Jensen comments about $MRVL $1T company around networking/connectivity/interconnects? - “So as the TPUs continue to accelerate, there’ll be pressure overall on margins. But the connectivity side, the AI networking side of the business has very rich margins” “Demand for … networking is simply insatiable” Also very positive read through as well for the $LITE and the other players. But for TPU margins it goes down at scale, which is understandable. - “they are placing orders in fairly huge demand, which basically gives us a lot more visibility.. runs all the way to 2028 right now” positive read through on overall AI demand since it’s 2026 now… and orders are out in 2028 - The initial order for 1 gigawatt, which includes XPUs and our networking has been received and will start Delivery in the second half of 2027. for our other two customers, we expect shipments to begin late 2026 and accelerate into 2027. $META custom AI program h2 2027 timelines - “Our revenue, our content per gigawatt will increase. you start putting a lot, you start putting embedding CPU cores into the same XPUs and making those chips basically multi die with lots of hvm.” Just for the GW modelers. - “For OpenAI we have delivered silicon and we are on track for production late 2026” OpenAI custom program timeline - “If you ask about 27 or 28 that will continue to grow. We expect in fact 28 to be a substantial growth from what we are forecasting in 27.” More about the demand ramp, go brrr - “Google, that we expect a diversity of sources from them” Mediatek (2454) primary beneficary, maybe $MRVL. Already expected though Google doesn’t sole source so they don’t get bottlenecked. There’s quite a lot of AI demand visibility way until 2028, which is bullish on the AI sector as a whole. Regardless, Broadcom ends the week +0% lol. TLDR: Strongly bullish AI demand, especially networking. Stocks don’t move in a straight line up, but demand curves 2026-> 2027 -> 2028.中文 · 翻译随手记几条 $AVGO 财报电话会议的零散笔记 - 营收目标重申（2027 年 1000 亿美元以上，市场明显是希望这次财报能上调的，所以股价跌了） 还记得 $NVDA 黄仁勋说过，$MRVL 会是一家围绕网络/连接/互联的 1 万亿美元公司吗？ - “随着 TPU 继续加速，整体利润率会有压力。 但连接这一块，AI 网络业务，利润率非常丰厚。” “对……网络的需求简直是无法满足。” 对 $LITE 和其他玩家也是非常正面的连锁反应。不过 TPU 的利润率随着规模扩大而下降，这可以理解。 - “他们下的订单量相当庞大，这基本上给了我们更多的能见度……现在排单一直排到 2028 年。” 考虑到现在是 2026 年，而订单已经排到 2028 年，这对整个 AI 需求是个正面信号。 - 首个 1 吉瓦的订单——包含 XPU 和我们的网络——已经收到，将在 2027 年下半年开始交付。另外两家客户，我们预计出货从 2026 年底开始，并在 2027 年加速。 $META 定制 AI 项目的时间线是 2027 年下半年。 - “我们的收入，每吉瓦的内容会增加。当你开始大量把 CPU 内核嵌入到同样的 XPU 里，让这些芯片基本上变成带大量 HVM 的多裸片结构。” 这是给那些喜欢算吉瓦模型的人看的。 - “对于 OpenAI，我们已经交付了芯片，并将在 2026 年底按期量产。” OpenAI 定制项目时间线。 - “如果你问 27 年或 28 年，那会持续增长。我们实际上预计 28 年相比 27 年的预测会有大幅增长。” 更多关于需求加速的信息，冲冲冲。 - “对于 Google，我们预计他们会有多元化的来源。” 联发科 (2454) 是主要受益方，可能还有 $MRVL。不过这个预期已经在股价里了——Google 不会只依赖单一来源，以免被卡脖子。 AI 需求的能见度一直延伸到 2028 年，这对整个 AI 板块是看多的。不管怎样，博通这周收盘涨跌持平，哈哈。 太长不看版：强烈看多 AI 需求，尤其是网络部分。股票不会直线向上，但需求曲线从 2026 年→2027 年→2028 年一直往上走。
- English · 原文@JonahK44 $NBIS is $META and $MSFT. $GOOGL has done a lot of Fluidstack deals with $CIFR to $WULF for more Colo. my guess is to plug in a lot more of their TPUS中文 · 翻译@JonahK44 $NBIS 就是 $META 和 $MSFT。 $GOOGL 跟 $CIFR 和 $WULF 做了不少 Fluidstack 交易，都是为了扩充 Colo。我猜他们是想接入更多自家的 TPU。
- English · 原文Ayar’s announcement today with Wiwynn is potentially very material for $SIVE regarding CPO -> rack scale deployments. As Wiwynn cloud clients include $AMZN, $META, $MSFT. And they’ve been in talks for $GOOGL TPU deployments. I think just for some reference architectures it’s around 512+ supernova light sourc a rack. So if $SIVE is the primary laser array supplier (which we expect, given Macom + Lumentum was removed from Ayar’s site). Even modest rack deployments would be very meaningful for revenue. This is just rack scale commercialization potential right now from $SIVE / Ayar / Wiwynn, which won’t show up in revenue financials yet.中文 · 翻译Ayar 今天和 Wiwynn 的声明，对 $SIVE 来说在 CPO -> 机架级部署这条线上，潜在意义非常大。 因为 Wiwynn 的云客户包括 $AMZN、$META、$MSFT。 而且他们一直在跟 $GOOGL 的 TPU 部署谈合作。 我觉得，单看一些参考架构，一个机架大概就要 512+ 个 Supernova 光源。 所以如果 $SIVE 是主要的激光阵列供应商（我们预期是这样，毕竟 Macom 和 Lumentum 已经从 Ayar 的官网上移除了）。 哪怕只是中等规模的机架部署，对营收来说也意义重大。 而这还只是 $SIVE / Ayar / Wiwynn 眼下在机架级商业化上的潜力，暂时还不会体现在财务营收上。
- English · 原文@EndratoxHatesW Really doubt anyone is quitting $RDDT to join FB's Forums.中文 · 翻译@EndratoxHatesW 真的不信有人会弃了 $RDDT 去加入 FB 的论坛。
在其他光模块产业链标的上也有持续观点
META 之外，以下标的同样收录于其观点库：