Spike Wang

@abao1130 @wangchangfu88 +1 画的贼好

@wangchangfu88 流程图可以试试让Codex 画 mermaid 图，更省 token

用 Codex 画流程图 简直夯爆了！！！

@william40152988 @imwsl90 难道不是自然语言嘛

@imwsl90 如果按第一性原理的话，最回归本质的是哪一个语言呢？

vibe coding 我建议先学习 javascript而不是 python javascript这门语言似乎什么都能做

@imwsl90 都vibe了还学锤子，直接自然语言编程了兄弟

@guansi minimax不知道该给你点赞还是点踩🤣

最近发现了一个特别有意思的用法。我把 GPT 和 DeepSeek V4 Pro 当高级工程师用，负责写 SOP、写 CI/CD 流程、设计方案。然后把这些东西一股脑丢给 Hermes Agent 里的 Minimax M3 去执行。 正常情况下，ChatGPT 一个小时能搞定，DeepSeek 两个小时也差不多了。M3 呢？已经干了一上午，四个小时过去了，还在和 CI 测试死磕。日志刷得飞快，错误改了一轮又一轮，就是过不了。 但神奇的是，我居然没有生气，反而越来越期待。 因为我突然意识到，M3 最有价值的地方可能不是干活，而是验收文档。 如果 GPT 写的 SOP，DeepSeek 写的流程，最后连 M3 这种级别的“小白员工”都能一步一步照着执行成功，那说明这个文档是真的写明白了。反过来，如果 M3 卡住了，大概率不是它一个人的问题，而是流程里还有默认知识、隐藏前提和作者自己都没意识到的经验依赖。 以前测试用户手册，要找新人。 现在测试用户手册，直接找最笨的 Agent。 某种意义上说，M3 已经从生产工具变成了质检工具。 它就像公司刚入职的新同事。能力可能一般，效率可能不高，但凡是文档里写得不清楚的地方，它一定能精准踩坑给你看。 AI 时代最好的测试工程师，未必是最聪明的模型。 有时候恰恰是最笨的那个。

@ModengSir 这两天确实比较明显

现在codex额度怎么这么不耐用啊 一会没看，额度没了？ 就做了一个小小项目 小伙伴们，你们也这样吗？🤔

@liuchang8877 @arkuy99 我要笑死了还有tibo雷达

@arkuy99 codexradar.com 有个检查是否降智的

codex 严重降智。

@thsottiaux GREAT! But what's better is to give pro-lite users more 5x usage than plus and made up to 10x plus total.🥰

Now that you can bank usage resets in Codex. Are you a hoarder or do you use them without breaking a sweat? How do you think about them?

@thsottiaux Hoarder +1

@kimikkko1487331 @antiAIvo 那你把我大minimax的营销能力放在哪里？

@antiAIvo 信的人这辈子都有了，智普在营销上我认为超越fable十倍不止

GLM 5.2 真的接近 Opus 4.8 的水平？ 有懂AI的 能评价下吗？ 是从哪个维度评测的？

@ChinaEastern_CN 中国东北有很多日本战争遗孤被中国人养大，这说明中国人始终关注的是人品而不是其他的东西

我妈经常跟我说“不要和中国人结婚”。 但我觉得用国籍来判断一个人适不适合结婚，真的有点太片面了。而且她一直反复提这件事，我也有点听累了。 想问问中国的朋友们，你们父母也会跟你们说“不要和日本人结婚”吗？

@thsottiaux @MehakdeepK81 Hit a bug with resets? I used my initial limit and invited a friend (they downloaded & messaged Codex). But my reset count didn't update. I thought I'd have 2 resets now (1 Tibo + 1 invite). Even after manual reset, nothing changed. Do invites overwrite Tibo's reset?

@MehakdeepK81 I would 1) Try Codex for free, it's included with any ChatGPT account 2) Realize it's awesome 3) Smile 4) ... 5) Smile some more 6) Get a subscription 7) Invite friends and get many resets 8) Keep smiling

Developers, you have $100. What are you buying ? -Claude -Codex

@thsottiaux When does 5.6 come up?

Hi, I'm Tibo and I just discovered Codex. AMA

@realsigridjin @AnthropicAI @realDonaldTrump so true

@AnthropicAI banning mythos/fable models for non us citizens is super wise move to sell more trump cards @realDonaldTrump

The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees. The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance. Access to all other Claude models is not affected. We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible. Read our full statement: anthropic.com/news/fable-myt…

@thsottiaux why my invite options appeared just yeasterday and today after update my codex-app, it disappears?

Heard your (amusing) feedback that it was at times annoying to receive a reset of your Codex usage without warning. Next time we press the button you will get to choose when it actually applies. Happy codexing.

@jinchenma_ai 那不然还能怎么vibe？

在咖啡馆看到有人搞 Vibe Coding： 1. 不用语音转文字 2. 不用muti agent 3. 不让 AI 连续深度思考 3 小时 4. 不在 Codex 和 Claude Code 之间来回切换 就只是输入提示词，然后盯着屏幕干等回复😂

@adamcohenhillel indeed

codex: implementing this will take 2-3 weeks also codex after 10 minutes: done

@SqSehrish 41

Only a genius can find the pattern in 10 seconds Can you?🤔

@n1ckstr3 @nvidia minimax-m3 is garbage

@nvidia open models are all the same change my mind x.com/thehypedotnews…

Introducing NVIDIA Nemotron 3 Ultra. A frontier smart open model built for long-running agents that need to plan, reason, use tools and keep working across complex coding, research and enterprise workflows. Up to 5x faster inference and up to 30% lower cost for agentic tasks. Learn more: nvda.ws/4x9nGps

@tun2049 why not ?

@kittyss02 学点经济学基本原理吧，球球你了

我还是不理解 作为全球第二大经济体 为什么工作遍地都是3000～4000