切捨魔剣時
5.7K posts


🚨Breaking: Microsoft just broke the AI hardware game.
They dropped bitnet.cpp — 1-bit AI that runs 100B models on your CPU.
No GPU.
82% less energy.
100% open-source.
This isn’t an upgrade.
It’s a takeover. ⚡
Repo: github.com/microsoft/BitN…
English

@ivanburazin I did this as a vibecoding blueprint. You can vibe your own
github.com/kristopolous/C…

English

After the Claude Code source code leak, a former PM extracted its multi-agent orchestration system into an open source model agnostic framework.
He studied the architecture, focused on the multi-agent orchestration layer (the coordinator that breaks goals into tasks, team system, message bus, task scheduler with dependency resolution), and reimplemented these patterns from scratch as a standalone open source framework without infringing on Anthropic's code.
The result is what @JackChen_x calls an "open-multi-agent." Unlike claude-agent-sdk, which spawns a CLI process per agent, this runs entirely in-process and can be deployed anywhere (serverless, Docker, CI/CD)
Check it out: github.com/JackChen-me/op…

English

@Radha_AI @Shruti_0810 It's been around for like 2 years. Someone used the word "claw" for it and now you're gaga over it ... alright.
English

🚨Claude Code just became a 1-person AI army.
32 agents.
5 modes.
3–5x faster.
Type ONE word:
→ autopilot = builds everything
→ ralph = won’t stop till done
→ eco = saves 50% cost
→ plan = thinks before coding
Auto model switching.
Auto resume.
Zero effort.
This isn’t a tool anymore.
It’s an AI OS.
Repo: github.com/Yeachan-Heo/oh…

English

@Shruti_0810 All of these features widely exist in other harnesses and have for over a year. These aren't new in the slightest. Did they need a cartoon character? Is that the difference?
English

@NoahEpstein_ You can do that in a few minutes of typing. The entire codebase has been abstracted and made modular as skills and a PRD.
github.com/kristopolous/C…

English

the fastest growing repo in github history is called "claw-code."
136K stars. 102K forks. in under 48 hours.
i run OpenClaw.
but here's what actually matters:
the community didn't just mirror anthropic's source. they rewrote it in rust, added better tooling, and shipped it faster than most companies ship a feature.
that's what happens when you accidentally open source a $2.5B product.
and if you want to actually USE what's in that source right now - the 3-layer memory system, the multi-agent coordination, the persistent background execution - you don't need to build it yourself.
openclaw already runs it. in production. on your machine, you just need to ideate how to actually take it to the next level.
the code is out there. most people will read about it and move on.
a small group will actually change how they build.

English

@Harish_521 You can do it in a few minutes of typing and about an hour of waiting...
github.com/kristopolous/C…

English

say hello to free-code
claude code source code fully recompiled, telemetry stripped, security guardrails prompts stripped, all working experimental features enabled
including ultraplan mode - a new async agentic planning mode where claude starts a multi-agent research session in the browser lasting between 10 and 30 minutes
since i know there are gonna be DMCA strikes on this i've uploaded it on the blockchain on IPFS

English

@support_huihui Not everyone knows how to hype
github.com/kristopolous/C…
English

Due to the leakage of Claude Code, the following projects were generated:
github.com/instructkr/cla…, Star 94.3K, Fork 88.5k
github.com/claude-code-be…, Star 5.7K, Fork 7.2k
github.com/paoloanzn/free…, Star 2.4K, Fork 930k
github.com/NanmiCoder/cla…, Star 1.8K, Fork 2.4k
github.com/oboard/claude-…, Star 1.7K, Fork 2.5k
English

I rode California's new Gold Runner train. I hopped off before the end. sfgate.com/travel/article…

English

@itsCathyDi My friend and I are still in pending. He's super into Minecraft, works on it all the time. His name is Logan and I'm Chris McKenzie
English


groups.google.com/g/aa-new-model i just made a Google group if you want to subscribe
it parses artificial analysis's evals hourly and then fires off an email if a new one hits the list so you can get an email when a new important model gets released
That's all it does.
English










