Sabitlenmiş Tweet
CCTeam把你已经在用的 code agent(Claude Code、Codex、Grok、Kimi 等)编成一支团队——任何会话都能跨厂商、跨机器spawn、派活、收结果,而你从Telegram、飞书或浏览器里统一指挥。
github.com/firstintent/cc…
中文
FirstIntent
275 posts
FirstIntent
@firstintentdev
Agent builder. Currently building ccteam — a control plane for one person to manage 50 AI agents.
Katılım Haziran 2021
594 Takip Edilen60 Takipçiler
@op7418 我也分享下自己生产力工具ccteam，用这套工具日常驱动50+ agent
github.com/firstintent/cc…
中文
Kimi serves as a vital member within the CCTeam agent cluster, working alongside Claude Code, Codex and Grok. This platform lets you spawn and assign tasks to these cross-vendor AI agents across different devices, gather all task feedback easily, and command the entire collaborative team conveniently through Telegram, Feishu or a web browser.
github.com/firstintent/cc…
English
Releasing the model weights and technical report of Kimi K3.
Kimi K3 is our most capable model: a 2.8T MoE model with native visual understanding and a 1M-token context window.
New model architecture: 2.5x the intelligence per unit of compute, not just more params.
Alongside Kimi K3, we're opening up more of the stack behind it — high-performance attention kernels, MoE communication library, and infrastructure for running agent environments at scale.
Model weights: huggingface.co/moonshotai/Kim…
Tech report: github.com/MoonshotAI/Kim…
Tech blog: kimi.com/blog/kimi-k3
English
fable5 规划调度，2 个 claude workflow 做架构设计，4 个 codex 编码执行，1 个 grok 调研。telegram 远程控制，这是我日常使用 ccteam 的场景。
github.com/firstintent/cc…
中文
Mountain Green, a new color for Xiaomi SkyNomad Series.
Inspired by the blue-green horizon of distant mountains. Warm, muted, refined, and quietly premium.
Finely milled blue-green pigment and metallic particles move naturally across the surface. As the light shifts, they reveal different layers of blue-green texture.
What do you all think?
English
明天开始 X 上预计会出现大量称赞 Kimi K3 的帖子。
因为可能由于某些原因从来没用过 Fable5(A 社太坏)，或只用过 GPT 系列模型，第一次使用 K3， 会有一些小震撼的，仅限 Max thinking effort.
中文
@gakonst That’s also why Grok blew up. it’s just super fast.
English
the harness must be always network latency bound unless it's doing some heavy tool call, anything else is just unacceptable - i can't believe that people accept harnesses taking 2 whole seconds to start
English
Grok 4.5 is a solid workhorse
Tim Sweeney@TimSweeneyEpic
Grok 4.5 is very good for vibe math in the field of Programming Language Theory. It's keeping up with Sol 5.6 and is way more fun to use than Fable/Opus 5 which love to spend hours creating superfluous documents (Claude code) or constantly nags me to "Continue" (web).
English
The #1 thing I want from any model right now: speed
Intelligence is already good enough. But waiting 1-5 min per task is the worst possible window: too short for deep work, too long to just stare at the screen. So I end up scrolling X
Agents are making all of us more ADHD
English
@runes_leo 试试 ccteam，我用了 3 个月了，Claude 调度 codex、grok 去干长周期的任务，再也不用多个窗口不用复制粘贴了
中文
确实。盯着屏幕看 AI 干活的时候，如果没有另开窗口去跟 AI 多线程干别的活，很容易焦虑——那 1 到 5 分钟最难受：既不够做一段完整的深工作，又长到让人忍不住去刷点别的。
我自己的解决方式是：有的窗口持续推进需要深入的任务，有的窗口就干零散、偏闲聊或琐碎的事；再用多个不同的模型入口和窗口、多条业务线并进，把等待时间填上。
但更值得警惕的是 ADHD 这一点。用久了会发现，不是你主动选择多线程，而是大脑慢慢被 AI 的节奏训练了——不知不觉就习惯了频繁切换、并行开线、很难再安安静静盯着一个窗口等结果。工具在加速产出，也在重塑注意力。👀
Zara Zhang@zarazhangrui
The #1 thing I want from any model right now: speed Intelligence is already good enough. But waiting 1-5 min per task is the worst possible window: too short for deep work, too long to just stare at the screen. So I end up scrolling X Agents are making all of us more ADHD
中文