Juan

[Teknium 🪽 @Teknium]

Hermes Agent can now control your windows or linux GUI Desktop apps - With ANY model.

[GitHubDaily @GitHub_Daily]

又发现一款实用的开源终端工具 Muxy，主打轻量低内存占用，同时塞进了不少实用功能。

按项目排列工作区，每个项目的标签页、分屏状态都能自动保存恢复，切换项目时无缝衔接。

GitHub：https://t.co/6lmJG4GsUA

内置了 Git 面板，能直接查看差异、提交记录、管理分支，还有文件树、全局搜索、Markdown 预览和轻量代码编辑器。

还能实时追踪 Claude Code、Codex 等 AI 工具的 Token 用量和费用，方便控制成本。

还有 iOS 和 Android 配套 App，可以用手机远程操控 Mac 终端。

如果你日常用终端跑 AI 编程工具，又想要一个不臃肿的项目管理环境，值得试试。

[Lou @louszbd]

huge thanks to @UnslothAI, GLM-5.2 can now run on a 256GB mac!

[Juan @thejuan]

Muxy > cmux

https://t.co/avDRKWpm5L

[Greg Barbosa @gregbarbosa]

Apple didn't, so I did: I made it dead simple to run macOS 27's local and Private Cloud Compute Foundation models in any app that accepts an OpenAI API URL.

Here's fm-proxy. A drop-in replacement, with no extra servers, and no extra keys. It's great in a minimal Pi setup!

[Ryan Stephen @Ryan__Stephen]

playing with realtime diffusion ui

[Nex @NexEcosystem]

The Rio 3.5 model broke the internet this week. The plot twist? It’s essentially our open-source model, Nex N2 Pro, wearing a different hat.

🤯 We analyzed the weights, and the recipe is exact: Rio 3.5 ≈ 0.6 * Nex N2 Pro + 0.4 * Qwen 3.5

It even literally introduces itself as "Nex N2 Pro" if you ask it without initial system prompt!

😂 We are flattered that the City of Rio used our work to achieve SOTA performance. Thanks for the ultimate benchmark validation.

🤝 But in the open-source world, attribution matters.
👇 Full mathematical proof & verify script in the first reply!

[Chubby♨️ @kimmonismus]

Wait what? Rio 3.5 Open 397B, developed by IT company of Rio de Janeiro's city government is now SOTA open source and even outperforming Qwen 3.7?

What is happening today.

Never heard of them before.

[𝗭𝗲𝗻 𝗠𝗮𝗴𝗻𝗲𝘁𝘀 @ZenMagnets]

Alibaba Qwen3.7 slowly fading into irrelevance at the frontier due to proprietary stance.
In it's place we have Minimax M3 and... *checks notes* Rio 3.5 397b, made by the municipal IT company of Rio de Janeiro's city government.

https://t.co/JgIJYVhoEi

[Juan @thejuan]

The US has turned hostile to all non-citizens, without realizing that a significant portion of technological advances and research, if not most, comes from that talent pool.

Extremely myopic move.

[Andrew Curran @AndrewCurran_]

According to Grok, Andrej Karpathy is an EB-1 extraordinary ability green card recipient, not a US citizen. Thus under these new restrictions he is not permitted to use, or work on, Mythos 5 or Fable 5 as of 5:21pm tonight.

[Andrew Curran @AndrewCurran_]

According to Grok, Andrej Karpathy is an EB-1 extraordinary ability green card recipient, not a US citizen. Thus under these new restrictions he is not permitted to use, or work on, Mythos 5 or Fable 5 as of 5:21pm tonight.

[Juan @thejuan]

Opus 4.8 max effort now feels lightweight on token use compared to Fable

[Xenova @xenovacom]

I gave Fable 5 one job: write custom WebGPU kernels for Gemma 4 inference.

It climbed to 84 tok/s, then hit a wall, insisting further optimization was impossible.

Hours later, Anthropic rolled back invisible LLM development safeguards, and it hit 255 tok/s.

The next day, access to Fable 5 was suspended globally.

[International Cyber Digest @IntCyberDigest]

‼️ UPDATE: It just doesn't stop: Almost 900 Arch Linux packages infected now.

https://t.co/xpRTsucoxA

[How To Prompt @HowToPrompt__]

Apple just did something nobody expected.

They turned 2 billion iPhones into local AI machines.

They open-sourced coreai-models, the entire toolkit that lets you export any HuggingFace model and run it natively on iPhone, iPad and Mac with zero cloud.

→ Runs 100% on the Neural Engine
→ No cloud. No API keys. No subscriptions.
→ Fully offline. Your data never leaves the device.

It even ships with skills for Claude Code, Codex, and Gemini, so your coding agent already knows how to use it.

100% Open Source.

[jpark @jparkjmc]

new policy from anthropic: if you use fable/mythos, they collect your data.

no exceptions. not even for enterprise partners.

[elie @eliebakouch]

mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community

also the fact that this is un purpose not visible to the user is crazy

[Sam Iglesias @siglesias]

My WWDC design session this year is about prototyping with agents. Hope you find it helpful!

https://t.co/5cezye70Z1

[Awni Hannun @awnihannun]

It's very cool that Apple shipped a 20B parameter on-device.

You can't put 20B parameters in RAM at any reasonable precision. To make it work they are using pretty exotic architecture by today's standards.

A small model predicts from the query (or prompt) which experts to load from Nand into RAM. The key distinction from a typical MoE is that you do this once per query and then generate all the tokens with the same experts (instead of switching the experts for every token).