从过年到现在,准确说应该是从Opus4.6~4.7吧。4.6那一代很牛逼,4.7开始垮了,垮了之后加上封号,我就转战到了codex。因为那会儿4.7也很垮,codex虽然也不牛逼,但至少他俩在我眼里面是一个水平。加上Codex量大管饱,还有重置,而且还不封号,基本上从春节后一直到10月份一直用的都是codex。 直到这次Opus5.5的出现,就是俩字,牛逼,真的牛逼。牛逼在哪儿呢?牛逼在说人话,更智能更聪明。 为啥说人话很重要?因为现…

从过年到现在,准确说应该是从Opus4.6~4.7吧。4.6那一代很牛逼,4.7开始垮了,垮了之后加上封号,我就转战到了codex。因为那会儿4.7也很垮,codex虽然也不牛逼,但至少他俩在我眼里面是一个水平。加上Codex量大管饱,还有重置,而且还不封号,基本上从春节后一直到10月份一直用的都是codex。 直到这次Opus5.5的出现,就是俩字,牛逼,真的牛逼。牛逼在哪儿呢?牛逼在说人话,更智能更聪明。 为啥说人话很重要?因为现…

一位长期使用 AI 编码工具的开发者反馈,Anthropic 的 Opus 5.5 在"说人话"和复杂任务处理上明显优于 Codex 及 GPT-6 系列部分模型,但账号封禁问题迫使其花费数天寻找稳定使用方案。

This great conversation on the rise of AI interpretability with @eric_ho of @GoodfireAI is also available on Spotify, Apple Podcasts, @snipd_app and here on YouTube (like and subscribe!): https://t.co/7TSOLBCvuZ

This great conversation on the rise of AI interpretability with @eric_ho of @GoodfireAI is also available on Spotify, Apple Podcasts, @snipd_app and here on YouTube (like and subscribe!): https://t.co/7TSOLBCvuZ

风投机构 FirstMark 合伙人 Matt Turck 与 AI 可解释性公司 Goodfire 的 Eric Ho 进行了一场关于“AI 可解释性崛起”的对话,相关内容已上线 Spotify、Apple Podcasts、Snipd 和 YouTube。这场讨论之所以值得关注,是因为它把“看懂大模型内部…

Last week, I launched my AI course to help you build a personal AI system that takes work off your plate. 100+ new members have joined since then, and I’m grateful. Instead of making you watch stale videos or sit thro…

Last week, I launched my AI course to help you build a personal AI system that takes work off your plate. 100+ new members have joined since then, and I'm grateful. Instead of making you watch stale videos or sit thro...

AI 创作者 Peter Yang 推出的个人 AI 系统课程上线一周即吸引 100 多名新会员,年费 150 美元并将于 7 天后涨至 200 美元。这反映出 AI 教育正从"追工具、看视频"转向"抄工作流、建系统"。

For two weeks, start a design, deck, or doc in the Claude app, and the work that follows in that conversation uses 50% less of your usage limits. Give it a try with Claude Sonnet 5.5. It has a strong eye for design an…

For two weeks, start a design, deck, or doc in the Claude app, and the work that follows in that conversation uses 50% less of your usage limits. Give it a try with Claude Sonnet 5.5. It has a strong eye for design an...

Anthropic 在 Claude 应用中上线了 Docs、Slides、Design 三类创作入口,并推出为期两周的用量优惠:在某一对话中起草设计、演示文稿或文档后,该对话后续工作量只按 50% 计入用量限制,官方推荐搭配 Claude Sonnet 5.5 使用。

Applies automatically each time you create a design, deck, or doc, and runs through October 15 on Pro, Max, and Team plans. How it works and full terms: https://t.co/zEtM50m851

Applies automatically each time you create a design, deck, or doc, and runs through October 15 on Pro, Max, and Team plans. How it works and full terms: https://t.co/zEtM50m851

Anthropic 旗下 Claude 面向 Pro、Max、Team 三档付费用户推出 Artifacts 创作额度促销,用户在生成设计、演示文稿或文档时自动生效,活动持续到 10 月 15 日。这意味着 Claude 正用"更宽松的创作用量"来强化它在文档与演示场景的生产力定位。

The future is verification-engineering. Proofs, (e2e) tests, benchmarks, linters… Some tests will be deterministic, some agentic. This looks great. https://t.co/ZSwn1UEupp

The future is verification-engineering. Proofs, (e2e) tests, benchmarks, linters… Some tests will be deterministic, some agentic. This looks great. https://t.co/ZSwn1UEupp

Vercel 首席执行官 Guillermo Rauch 提出"验证工程(verification-engineering)"将成为软件开发的下一阶段,与此同时,一款名为 e2e 的开源智能体测试框架同步亮相,尝试把确定性测试和 Agent 驱动的测试混合在同一个工具链里。