iefox5

@kimmonismus come to Shenzhen. very close to Guangzhou

My first few hours in China, first impressions At the invitation of XPeng, I'm visiting China and starting tomorrow, I'll be attending several events to see the future of robotics and the automotive industry, and of course, learn quite a bit about AI. My plane landed around 4 PM; jet lag is definitely taking its toll. My first impression from the plane: construction cranes everywhere. China is really building on a massive scale. Second impression in a Chinese Uber: electric vehicles probably make up 90% of the cars here. Germany, as a "car nation," can't compete with that. EVs are the standard, not the exception. Third impression: Digitization everywhere. Everything is digital. As someone from Germany who still has to do a lot by mail, it was practically a culture shock. China is miles ahead when it comes to digitization. Everything runs via QR codes and scans. Fourth impression: At least in the big cities, the infrastructure is modern. Yes, there's a lot of traffic noise and congestion. But the roads are well-developed, the network is functioning, and it's modernly designed. Tomorrow I'll see much more. It remains exciting!

@sitinme 你的目的是什么？跑本地模型吗？具体什么场景，为什么不用API？

兄弟们128G 的M5Max 我冲了，后续给大家评测大模型！ 对比了下企业比个人（教育+国补）贵了差不多贵5000，纠结了半下午， 最后还是买了，不折腾了，要努力挣钱了💪！

@johnAGI168 GPT image 2 怎么拿到的阿?

两大SOTA模型强强联手👬 ChatGPT image 2 ➕ Seedance 2.0 🎬 制作广告简直是手拿把掐🤏 Created in Higgsfield with Seedance 2.0

@mronge Can I run it on Android devices?

Yesterday: iPad demo Today: iPhone demo Same headless Mac mini setup, but now you can manage AI agents from your pocket. Here’s a quick demo of how I remotely connect from anywhere with my iPhone to monitor what’s happening, and take over when needed

@nateliason @steipete glm 5.1 turbo

I have full faith that @steipete is going to make GPT in OpenClaw amazing... But the switch from Opus has been tough today. Any other models people are liking that are worth trying? Minimax 2.7?

@theo GLM V5 Turbo

I decided on OpenCode Zen btw

So, uh, what subscription should I be using for my OpenClaw now? 🙃

@LoopOnChain @steipete @davemorin yes GPT 5.4 always stops and asks you about an action you gave permission to 10 times already

@steipete to me the subtext here to be honest is that chatGPT 5.4 does not seem to perform nearly as well with Openclaw as Opus If it did, really not much of an issue for claw users i don't think this is just a me problem, many power user friends feel the same BUT My openclaw setup was designed and optimized around Opus Wondering if you have any tips or guidance on tuning openclaw to work better with 5.4

woke up and my mentions are full of these Both me and @davemorin tried to talk sense into Anthropic, best we managed was delaying this for a week. Funny how timings match up, first they copy some popular features into their closed harness, then they lock out open source.

@ewind_dev VPN 是不是不行

我是傻逼 🤡 我居然一直没用 tailscale ssh，只要一次性 sudo tailscale set --ssh 开通后，其他所有机器不用人工配私钥和 ssh config，直接 ssh user@device 就能让本地 claude 进去狠狠造了 接下来就是串流看 computer-use / browser-use 了！

@jxnlco @thsottiaux Please tell me the thing behind the computer is real.

Thanks @thsottiaux We reset rate limits so you can try out plugins!

@libukai 关键怎么不看呢？Feed 都自动出的

默默做了个决定，以后尽量不看中文账号的内容了，然后努力咬牙把 Anthropic 每篇文章都认真精读一遍。 虽然精华进，不一定精华出，但垃圾进，一定垃圾出。

@xiaojietongxue 怎么连接的API呢？

🔥通过Claude Skill调试脚本，然后用Seedance生成，我之前基本是用Gemini的Gem来完成这些任务，现在试一试Claude，发现它的理解能力和遵循力还是不错的。 👺感受一下这个高速运镜的效果！

@gkxspace @TopviewAIhq 能用api吗？

太狠了，Seedance 2.0 上线Topview了，Business年付365天无限用。 给AI一句话，它自己拆分镜、生成多场景、剪成完整视频。这事 @TopviewAIhq 的Agent V2做到了。 你不用自己写分镜，不用一段一段生成，不用开Premiere手动拼。给它一个想法，Agent自动处理整个流程，直接在时间线上成片。 而且时长不限。不是那种只能生成15秒片段的工具，3分钟5分钟都行。 能从头到尾做完一支完整视频。

@IshwarJha @Jacobsklug and why is this?

@Jacobsklug There are so many things in openclaw that it would take a lot of time for Claude Cowork to catch up. The biggest advantage is you get 15X more tokens than if you use Claude API to connect with openclaw.

I fully switched to the Claude Cowork train, over OpenClaw. Time for you to do the same.

@Jacobsklug how do you send messages between sessions? like let one session to send a prompt to another session?

@thdxr April.

where the heck is deepseek v4

@gustojs @theo @thdxr exactly

@theo @thdxr I mean, they paid for those tokens. Why not use them.

@binghe 已经开放API了吗？

火山引擎 Seedance 2.0 API接口公开后，视频输入与非视频输入的价格差异达 18 元 / 百万 Token！ 包含视频输入的价格为 28 元 / 百万 Token，仅使用图片、文字或音频输入的价格为 46 元 / 百万 Token。（如图1） 计价方式：视频输出按宽 × 高 × 帧率 × 时长计算消耗，720P 分辨率五秒视频约消耗 10.3 万 Token，对应一秒视频输入成本 0.58 元，非视频输入成本 0.95 元。（图 3） 官方案例显示，输入视频参考人物动作、镜头语言或运镜方式，即可按 28 元 / 百万 Token 计费。 图 2，就是视频输入。（里面有人物的动作） 我的理解是有视频参考模型运算更简单，不含视频的话，文本到视频是有一系列运算的，消耗更多的算力。 所以，自己找运镜镜头，或者自己拍一段动作，传上去，帮助模型节省 token ，这样价格就能降下来。 做精品漫剧，这个费用还是可以的！能接受！个人可能有点吃力。。 免费的积分你也可能薅羊毛！

@YunaisChen 我一直很疑惑，为什么你们总对自己迷之自信？当年ChatGPT 刚出来的时候别人也是这么说的，不如这个不如那个，MidJourney刚出来的时候我们公司的摄影师不屑一顾。你们这跟威尔史密斯吃意面有什么区别。

试用了半个小时的新stitch，评价是NPC。拿了两张Pinterest的landing作为参考图，对于我来说，stitch产出的东西没有任何想用的欲望。首先还是AI对文案全大写的执着没有改善，另外就是僵硬的layout，以及内容质量不够高。不清楚Google demo那种质量的设计稿是怎么来的。

@Dainer_Jun @TiezhuCrypto OpenClaw 定位的就是普通用户和通用 agent，真正的coder功能再全也不会用的

@yang_pei @TiezhuCrypto Clawbot的cron 也不行吧，cc 现在有loop + remote control + schedule . 基本上也没怎么需要太多设置，真的会基本cs 的肯定对cc 没问题👌

我用到现在，我没看出来有什么是Claude Code+Codex不能做，而openclaw能做的。