Ervin

@Trader_S18 退一亿步讲，这种情况下，强哥在肯定第一时间跑到老大身边。

目前来看，伊朗暂时是安全的。川普无意将枪击事件与伊朗联系起来。但是万斯可能就危了，领导没走你先走是什么意思🤡🐶🌚

@JJ_2206 @qq114154 @DecentralLife 贱不贱

@qq114154 @DecentralLife 就近原则，又不是哪里都要让LD先走

Ok，这次是真的，不是舞台剧。川总在跑的时候腿都软了，是被人架起来端走的。 如果是舞台剧，起码要把国旗安排到位，摆拍一分钟，振臂高呼fight fight fight。 太危险了，真的有人行使他的第二修正案来了。

@jxzdmzw 你太罗嗦，一句话就能说清楚：台巴子别给脸不要脸

现在中国大陆唯一还有耐心跟你们谈和统的一派，叫共产党。中共惠台十条发布，被大陆人骂到不得不关闭评论区。大陆如果一人一票，谁还留着台湾等过年？ 有时候我也真是佩服你们可以如此弱小，又还能如此傲慢。但凡有碟花生米，也不至于喝成这样。醒醒吧，你们有什么资格跟大陆提条件？有什么能力跟中共提条件？ 民主是什么黑科技吗？写个民主符咒挂身上，解放军看到立马缴械投降吗？ 虽然我对共产主义完全无感，但未来大陆怎么走，早就轮不到你们台湾人来指手画脚。 大陆给了你们一国两制的优待，不去干涉你们的政治制度，你们倒给鼻子上脸了，2300万要来决定14亿应该如何生活？

@Biantaiboooo 那堆片假显得很好看？

以后还需要摄影师和设计师吗？ ← iPhone 直出　 ChatGPT 优化 →

@vRnrijyokx5CVP9 @BeijingDai 呵呵

@BeijingDai 通化事件は三千名の日本人が、中国共産党軍と朝鮮人ゲリラによって虐殺されるという惨劇 しかもその事件の写真を南京虐殺の証拠写真としてでっち上げる。 左翼と共産主義は嘘を平気で使うなよ！

After Japan surrendered in 1945, a group of Japanese nationals in Tonghua, Northeast China, launched an armed uprising. They killed CPC personnel and many wounded soldiers. The rebellion was eventually put down by Communist forces together with the Korean Volunteer Army, and several thousand Japanese involved in the uprising were executed. This is known as the Tonghua Incident. The Japanese like to use the Tonghua Incident to reinforce their victim complex. In reality, the core reason Sino-Japanese relations remain so poor today is precisely that incidents like Tonghua were far too few. If, in 1945–1946, China had witnessed thousands of similar events — helping those Japanese who so coveted Chinese land stay in China forever — I believe Sino-Japanese relations today would be much better than they are. History, however, allowed those millions of Japanese war criminals to return to Japan. Their crimes never received the punishment they deserved. Worse still, their militarist ideology continued to spread throughout Japanese society. As a result, a hundred years later, China still has to expend effort to root it out on the Japanese islands. In fact, for the Japanese civilization, having all these people stay permanently in China back in 1945-1946 would have been the best possible outcome. I have many Japanese followers and I believe more and more Japanese will agree with my viewpoint, especially when Chinese bombs fall on their heads in the future.

@susancrabtree 狼来了喊太多遍，人们就不关注狼了，关注的是谁总在喊。

🚨🚨 SECRET SERVICE EXCLUSIVE: For the two weeks before the shooting at the White House Correspondents' dinner, the Secret Service had been investigating a mystery shooting that took place near the White House the morning after Easter Sunday around 1-2 a.m. The White House/President Trump has been demanding answers from the Secret Service - but the USSS has been at a loss. The agency found rifle casings at K and I Streets, and had the license of the car, but no images of the shooter, and the license was stolen, according to multiple Secret Service sources. The case had gone cold — to the frustration of the Secret Service and the White House.

总理为外交部长开外交部长国家的新能源车。

@dotey 陈博渊，这名字太棒了

如果有 GPT Image 2 的意见，可以在原推评论反馈，Boyuan 是 GPT Image 核心团队成员

@PatrickZhengCHN @En_chinaNews 哥们这是AI

China breaks the world record with a spectacular show featuring 65,000 drones in Shanghai!😘🇨🇳 Credit: @En_chinaNews

比亚迪的新发布的车太帅了！ x.com/XueJia24682/st…

@weiyux2021 这道题这么出：如果必须消失一个AI公司，你选哪个？

好奇，这几家AI巨头，未来会是谁吃掉谁？

@brucexu_eth 哈哈，有人因为claudecode+deepseek 没有调用他开发的skill就把DS评价地一文不值。怎么说好呢😄

刚做了一下测试，我觉得这个真的是有点牛逼了。我感觉它的水平跟 跟 ChatGPT 5.5 Pro 的水平接近，比 5.4 Pro 要好一点。5.5 Pro 这可是最新的模型啊。我发给了它一个比较长一点的 prompt，然后要求它输出，它的结果实际上是比 5.5 Pro 好，好的点在于什么呢？在于第一个，它速度要比 5.5 Pro 快好几倍。其次，内容层面的话，它精确捕捉到我是 Crypto 友好的，所以它在这个投资策略上，会更加激进地帮我做了 Crypto 的配置。而那个 5.5 Pro 它虽然拥有我的那么多 context，但是还是给我配置的是那些传统的，而且把 Crypto 配置设置为高风险、实验性的尝试。我去，感觉这个是真的有点牛逼了。 有人认识 Deepseek 团队的人吗？或者是 Deepreal 开发者社区相关的都可以。

@Linahuaa J20对比的是F22

Chinese genius consistently beats Western expectations >High-Speed Rail: In the early 2000s, foreign experts said China would take 20+ years to master 350 km/h trains. By 2008, they had the world's longest high-speed network, built largely with indigenous technology. >5G & Telecoms: Huawei was a telecoms follower in the 2000s. By 2019, it was the undisputed global leader in 5G infrastructure, owning critical IP that U.S. allies could not bypass. >Beidou Navigation: The U.S. GPS monopoly was considered unassailable. China launched Beidou's global system in 2020, achieving centimeter-level accuracy with a fully homegrown satellite constellation. >J-20 Stealth Fighter: Western analysts in 2005 predicted a first flight "no earlier than 2020." It flew in 2011 and was combat-ready by 2018, often compared favorably to the F-35. >DUV Multi-Patterning to 5nm: When ASML's EUV was denied, SMIC was supposed to be stuck at 14nm. Instead, by 2025, it was producing 5nm Kirin chips using quadruple patterning—a feat many Western engineers thought too complex for yield scaling. >LDP Gamble: China didn't try to copy ASML's LPP approach, widely considered the only viable path. They chose LDP- a route that, if successful, could actually be simpler and more power-efficient. What does that tell us? >EUVs will surely come earlier than expected (Western doomsayer forecast was made before AI explosion- which only helps China accelerate) >Chinese AI chips will come earlier than expected. DeepSeek would already be the #1 lab in the world if they had compute. And they WILL get compute earlier than expected. >China will lead in AI. China will super-lead in practical AI implementation on a societal level (they already mog at this) >China 2.0 will drastically widen the already large futurism gap before US politics can even figure out how to re-distribute the AI spoils to all the jobless broke niggas High chance that by 2050 China will be the undisputed #1 megapower, while US will become SovietUnion 2.0 Time to learn Mandarin, niggas. Jensen tried to warn you!

@chaozuoye 如果可能离A公司远点。Codex+GPT5.5+Image2，这是目前最强的组合，还无网络问题，无kyc问题，无恶意封号。 可以选择的话，远离A公司，选择OpenAI

到目前为止，这是我给所有想用好 AI 的人最重要的建议： 1. 给 Anthropic 付一点钱 2. 下载并打开 Claude Code 3. 按 Shift-Tab，直到出现”plan mode on” 4. 打开手机语音备忘录，把你想做的事全部说出来。觉得说完了？继续说，至少录 10 分钟，越长越好 5. 把语音发到电脑，用飞书妙记转录成文字 6. 在 Claude Code 里输入：「我从来没用过你，但我说了一些事情，粘贴在下面。请读完后问我问题，帮我搞清楚怎么用你做出一些厉害的事。一直问，直到我说结束为止」 7. 粘贴转录文字，回车然后就把主动权交给 Claude。如果这个方法有效，欢迎来找我说说。 如果这听起来很奇怪，就把这段话直接丢给你在用的任何 AI，然后说：「有人让我把这个发给你，我想试试，但不知道怎么开始，你帮我。」 好奇心和坚持，比什么都重要（来源于Matt Stockton）

@dynamicwangs 对的，一些图像非常像SD1.5的伪影和噪点。另一些则干净美观。

都在夸gpt image2 我来唱唱反调吧， gpt image2 应该是学习了大量的AIGC内容，导致其画面的高频信息经常异常的溢出，在一些场景下人物面部特征很像sd1.5时代的模型出图，肢体有残缺，在雨夜的这张环境下还有类似过拟合产生的异常噪声，也很容易出现类似第三张图的物理性差错，看来gpt image2 虽然在编辑和理解能力上有了大幅提升，但AI模型“近亲繁殖”导致的畸形问题已经开始暴露出来了。

@En_chinaNews 哥们，你这是AI

@Lostlanding 三方面的原因：第一老美不希望看到亚洲团结；第二小日子认为二战败给美苏不是中国；第三小日子认为蒙古族满族（姑且不论真满假满）统治过我小日子族为何不能。 三层一层比一层深。 只有让断了第三层念想，它们才能被制服。 只抓第一第二层，“贼惦记”的勾勾心怎么都会死灰复燃。

【日本政要不参拜靖国神社是会死吗？】 靖国神社里供奉的不只是普通战死士兵。1978年起，包括14名甲级战犯在内，被合祀进了靖国神社，东条英机就在里面。中韩的反对的是“不能把战犯和战死者混在一起供奉，然后由国家领导人正式参拜”，而日本故意模糊了这条线。德国有没有这个问题，因为德国明确区分了普通士兵的墓地是墓地；纳粹战犯是罪犯，没有供奉在任何国家纪念场所。 日本有更合适的战争纪念场所吗？ 有。千鸟渊战没者墓苑供奉无名战死者，没有战犯，各国都接受。天皇从不参拜靖国神社，参拜的是千鸟渊。但千鸟渊门可罗雀。 参拜靖国神社是一个主动的政治选择，不是无法避免的宗教义务。 从纯粹的国家利益角度看，日本的选择是损人损己的，因为靖国神社参拜的外部代价非常大。 对华关系： 中国游客减少，日本旅游业直接受损。 中国市场对日本产品的抵制，汽车、电子、食品。 中日高层外交冻结，战略沟通渠道关闭。 中国加速推进东海巡逻，钓鱼岛现状持续被改变。 对韩关系： 韩国对日本产品的消费者抵制。 韩日安全合作受损，而日本极度需要韩国配合应对朝鲜威胁。 韩国在国际场合对日本的支持度下降。 对东南亚： 日本在东盟的软实力持续损耗。 东南亚华人社区对日本品牌的负面情绪。 这些代价，每一个都直接打击日本的经济命脉。 为什么还要参拜？ 第一，国内政治的需要 靖国神社参拜在日本国内有固定的政治支持群体。日本遗族会和右翼保守势力的选票意义重大。对日本政客来说，参拜靖国神社是低成本、高回报的国内政治操作。代价是中韩抗议，但中韩又不能投票。参拜靖国神社，本质是政客牺牲国家利益换选票。 第二，历史修正主义的信号 参拜靖国神社，在日本右翼的叙事里有更深的含义：战争是为了亚洲解放的正义战争；战死者是英雄，不是侵略者；甲级战犯是被胜利者强加的不公正判决的受害者。参拜是对这套历史叙事的背书，是有意识地反对二战秩序。 第三，对中韩反应的系统性低估 2012年之前部分成立，中国的反制能力有限，经济代价不够大。但2012年之后这个计算彻底失效了，中国的反制是真实的、持续的、成本越来越高的，所以日本人开始抱怨“中国脾气太大”。 参拜的收益——国内保守票仓的支持，直接转化为选票。 参拜的代价——中韩的抗议、东南亚的不信任，不转化为日本国内的政治成本。 在这个激励结构下，理性的日本政客会继续参拜，因为收益大于可见代价。 靖国神社真正的问题是责任和代价的错位。参拜的政客获得了选票，承担代价的是日本企业、日本外交官、和日本普通百姓。

@cccchuizi 不同版本不一样？还是我没有登录？

Claude Desktop 现在支持第三方 API 了（可能早就支持，只是最近才被发现）。 L 站的佬友们真的太聪明了 原理很简单： 1.打开 Claude Desktop，先不要登录 2.打开菜单 → Help → Troubleshooting → Enable developer mode（启用开发者模式） 3.启用后会出现 Developer 菜单，进入 Developer → Configure third-party inference 4.填入你的 Base URL 和 API Key，选择 Apply locally 即可 配置完后就可以用中转站的 Claude 了，省 Pro 订阅的消耗。 原文：linux.do/t/topic/2032192 官方教程:#h_c00b8c02e0" target="_blank" rel="nofollow noopener">support.claude.com/en/articles/14… 大家可以试试，欢迎反馈补充。

@mikefutia CHAT

Claude Code + ChatGPT Images 2.0 is f*cking cracked 🤯 I rebuilt my static ad system inside Claude Code on the new ChatGPT Images 2.0 model. One brand name + one URL = 40 production-ready static ads. All inside Claude Code. Perfect for DTC brands and agencies who need high-volume ad creative without briefing a designer or spending hours in Canva. If you're finding winning ad concepts on Meta and manually recreating them one at a time — copying prompts, pasting product details, tweaking aspect ratios, downloading, organizing... This system eliminates the entire loop: → Give Claude a brand name and URL → It researches the brand's fonts, colors, packaging, and photography style → Builds a Brand DNA document from scratch → Fills in 40 proven ad templates (headline, us vs them, testimonial, UGC, review cards, stat callouts) with brand-specific details → Fires every prompt to ChatGPT Images 2.0 with your product photos as reference → Downloads finished ads into organized folders with an HTML gallery No manual prompt filling. No Canva templates. No copy-pasting between tools. What you get: → 40 ad formats filled with your exact brand colors, fonts, and copy → Text that actually renders correctly (the new model handles dense copy, logos, and multi-language callouts cleanly) → Product photos passed as reference so the model matches your real packaging → A reusable system — new brand, new folder, same pipeline Built 100% in Claude Code with ChatGPT Images 2.0. I put together a DIY playbook showing the exact architecture so you can build this yourself in Claude Code. Want it for free? > Like this post > Comment "CHAT" And I'll send it over (must be following so I can DM)

My wife collapsed in our hotel room in New York today “Call an ambulance, I can’t breathe” she was screaming My heart dropped If she ends up in an American hospital we are financially ruined I went on the Lufthansa app to book a flight back to Frankfurt, but unfortunately pilots are on strike today “Please I’m begging you” she was lying on the floor I sighed and called 911 She is now in surgery as apparently her appendix “almost burst” I am extremely scared This is going to cost us at least $100,000 She could have received much better care, for free, in Germany I will never visit this barbaric country ever again