服务器技术专家,曾就职于阿里云 ,有赞,外企,在线教育等; 👑 AI coding - 职业工程师 💻 只分享硬核AI编程工具,技巧等

中国
这个牛逼,把codex 的computer use 能力赋予给其他harness. 实测在我自己的xharness 里也可以使用codex 的computer use 了,速度和准确率的确要比cua-driver的要好用。 原帖是在claude code里使用codex的computer use, 提示词: ``` "Set up Codex's computer use as an MCP server for you on my Mac. Find the "cua_repl" entry in ~/.codex/plugins/cache/openai-bundled/unified-computer-use/<newest version>/.mcp.json and register it as a user MCP server called "codex-cu" with the same command, args and env. Then test it by using Calculator in the background to work out 12 × 12." ```
everyone thinks codex computer use only works inside codex it doesn't, it's a local mcp server in the chatgpt mac app and claude code can just call it, even headless with claude -p same 8-task test: opus 5.5 on codex's engine got 6/8, codex itself got 6/8, cua driver got 3-4/8 at 4x the cost cua's background mode on mac only goes through accessibility, so canvases and drags just don't land codex's engine sends real clicks and drags to the app without touching my cursor no hover though, and it's unofficial, so enjoy it until a chatgpt update breaks it give this to your claude: "Set up Codex's computer use as an MCP server for you on my Mac. Find the "cua_repl" entry in ~/.codex/plugins/cache/openai-bundled/unified-computer-use/<newest version>/.mcp.json and register it as a user MCP server called "codex-cu" with the same command, args and env. Then test it by using Calculator in the background to work out 12 × 12."
1
150
openAI 的模型已经烂到没人买Pro 500 来炫富了。 有钱人是有钱,不是傻,谁买 $500 的反而可能会被别人当成傻子,适得其反。 本来 $500 的套餐是体现身份和能力的炫耀价值,可惜它自挖坟墓,减半用量,$500 现在只相当于以前的 $250。
251
有推友反馈,他用kimi code 提交代码,会出现co-authored-by: Claude Fable 5 的记录,于是怀疑kimi 蒸馏了Claude. 我个人觉得是Kimi躺枪(虽然推友说他本地没有安装CC,我也许也错了),考一下大家,你该如何调试分析找出问题的root cause?
太乐了,Kimi是不是蒸太多Claude了,怎么把Git co-author也蒸进去了
2
296
推荐一个追踪模型是否「降智」的网站 Nerf Bench:以发布首日表现为 100%,然后每天持续复测质量+成本,低于 90% 算下降。 已覆盖 Opus 5.5、Sonnet 5.5、GPT-6-Astra、GPT-6.1-SOL。 bridgebench.ai/blog/how-nerf…
7
3
14
4,976
找到了读书的终极方法:把书做成一集一集的视频。 世人都没耐心看书了,注意力早被短视频训练成只看得进视频(悲)。 于是用Opus 5.5把书的几章做成视频,居然看得津津有味。先拿纳瓦尔的《How to Get Rich》试了试,效果很好,香!
14
49
229
21,315
假期了,放下AI,看到这个新闻真的心痛。
2
734
真的震惊,梅西到现在还不会英语,已经在美国踢球这么久了,所以说英语真的很难。
994
今天半夜凌晨2点,chatGPT 又要重置了!大家不要条件反射,坚决当做没看见,好好过节日! 10 月3日凌晨2点钟。
Global reset landing tomorrow 10am PST for all paid ChatGPT accounts. Apologies for the slow start with GPT-6.1 Sol, it's now back to running at expected speeds after the massive load spike in the first two days.
21
1
2,015
继 OpenAI 后,Cloudflare 也带着决策模型 Clef 和 Clef-flash 入场。 jev 还没火几天,百亿身家就要清零了,人生如幻啊! 当代码、软件、小模型都越来越容易复制,真正难复制的,似乎只剩超级大模型,软件创业只能小打小闹。
"Today, we’re releasing two Cloudflare-trained decision models, Clef and Clef-flash" blog.cloudflare.com/clef-dec… Never seen an idea spreading so fast.
3
7
2,925
openAI Dots 的第三方开源版本OpenDots 来了! 这个团队之前我转发过,它有几乎所有巨头bot 的开源版本: grok bot -> OpenBot (这个项目目前完成度最高,推荐) muse -> OpenMuse Dots -> OpenDots github.com/CopilotKit/OpenDo…
🎉 Introducing 𝙾𝚙𝚎𝚗𝙳𝚘𝚝𝚜 Self-hostable, always-on AI coworkers that works with ANY agent harness. Includes: - Computer use: browser, terminal & files - Bring agents to Slack, Teams etc - Spaces and Pages for projects - Voice calls - Web and Mobile Repo → github.com/CopilotKit/OpenDo… Powered by @CopilotKit and AG-UI. Clone this template and customize it however you want. Enterprise-ready.
1
7
1,046
有人真金白银买了 Manus 会员吗? 我搜了下中英文 X,除了发布那天的集体转发狂欢,这两天,几乎看不到任何回声。 做AI 应用太难了,强如manus 也翻不起浪花了,现在大模型巨头都有原厂替代品。
54
1
1,229
AI修图最大的噩梦:改一处,全图跟着变。 Ideogram 4.5 专治这个。这个小哥用它连续换了 100 次衣服——脸没崩、背景没飘、光影纹丝不动,只有衣服在变。 官方敢自称"最精准的编辑模型",靠的就是多轮编辑不累积漂移。感觉很适合精修照片、电商换款!
looks like they solved it. besides the fact that it reduced my head size by 20% at the 26th iteration and slowly made me fatter, everything else was almost intact. my tattoos are also a bit faded on the 100th edit, but no one in their right mind should be doing that many edits. very cool.
1
871
最好在国庆假期到期前在闲鱼买好Gemini Pro账号,过完国庆肯定要涨价了。 因为gemini 4 argon 就要正式发布了,并且看跑分图,agentic coding 能力特别优化过哦。
5
1
2,241
昨天anthropic 发帖说智谱的GLM-5.3 模型网络攻击能力很强,我不信,想试试。 找了一圈,终于找到abliterated消融安全层的版本API 在线提供商了。地址: abliteration.ai/pricing 拿我自己的网站练练手?学习下red teaming安全知识。
12
39
436
51,632
一段超级提示词,让 Opus 5.5 直接画出整个项目的架构地图。 和之前 Devin 的 deepwiki有比较大的区别:deepwiki 偏文档式的 wiki,这个偏可视化一些。 我拿自己的项目实测了一遍,效果非常好(看图)。 原作者的提示词是针对他的项目特化的(见原帖),我让 AI 改写成了通用版本,放在下面 👇 gist.github.com/linearuncle/…
had a crazy idea today for the past two months i've been building a pretty massive app and truth be told, i often play catch-up on architecture, because features move so damn fast so today i asked claude (opus 5.5, high) "I have a weird task for you. I want you to look at the app and create, for lack of a better word, a canvas that I can zoom in and zoom out and move around to see how our architecture actually works." (full prompt in first comment) this came out pretty great and it truly visualized, with steps and connected graphs what everything does 10/10 would recommend
9
32
161
22,417
Google 你是不是有病? Gemini 4 Argon 发布了,结果只给网络安全人员用,普通群众想摸都摸不到!你发个der 啊! 看到推友们转发,害我 antigravity 里找了半天没找到。
On behalf of Gemini 4’s modeling team to improve VibeCoding capabilities (spanning WebDev, 3D, games, full-stack, Android dev, and more), we just landed 91.9% on VibeCodeBench! As the AI race heated up, our release cadence accelerated dramatically. The pressure was immense - we worked day and night, non-stop, to land critical recipe improvements and push our model performance forward! Today, with Gemini 4, we came back strong! Huge thank you to my teammates who held the line through countless days and nights without ever complaining, and to our horizontal partner teams and leadership who believed in us and gave us unwavering support. Onward and upward!! 🚀
34
3
53
20,169
不刷信息流,重要内容自己送上门!让 dots(或bot, AI)融入我的体育迷日常!让AI 发挥主动性! 从今天起,dots每天帮我盯好这些: 🏀 NBA:新闻 + 每日战报 ⚽ 皇马 & 巴萨:西甲动态全掌握 🇪🇸🇫🇷🇩🇪 西班牙 / 法国 / 德国国家队 🐐 梅西 & C罗:新闻 + 比赛战报
1
2
981