Groktopus

@tonysimons_ Sadly hosting these on X means Hermes can’t scrape and I can’t have a conversation with my clanker about it.

Part 11 is live. 🪽 Hermes can run agents, tools, cron jobs, gateways, sessions, and entire multi-profile workflows. But once you’re operating all of that, you need a control plane. That’s the Admin Layer. Profiles create clean runtime boundaries. The API server connects custom frontends and teams. The web dashboard lets you inspect and manage the whole system without living in the terminal. - Switch profiles. - Edit skills. - Manage cron. - Inspect sessions. - Watch active work. - Check gateway health. One dashboard. Multiple agents. Total control. Only one part left. Hermes Agent Master Class Part 11: The Admin Layer👇

@ligma__sigma @Alibaba_Qwen We’ve got Temu Alex Jones here! As usual with these types, they believe what they want to believe. Even if the “evidence” has all the hallmarks of US gov propaganda. gist.github.com/magnus919/906b…

@Alibaba_Qwen NO THANKS! You can keep all the extra code vulnerabilities your Chinese AI model injects into non-Chinese users code products.

Qwen3.8 is launching and going open-weight soon!🌐 With a massive 2.4T parameters, this model is continuously evolving. We believe it’s one of the most powerful model available today, compatible to leading frontier AI models , second only to Fable 5. You don't have to wait to test it. Just now, the Qwen3.8-Max-Preview made its debut on Alibaba’s Token Plan, Qoder, and QoderWork. Be among the very first to try it out. Can't wait to hear what you build. Stay tuned! 🚀  Token Plan international：qwencloud.com/pricing/token-… China：platform.qianwenai.com/pricing/token-…

@ligma__sigma @TianhangZhuzth Share sources. Images aren’t sources.

@TianhangZhuzth Why did you create a Chinese AI model, Qwen, that created 130% more security vulnerabilities in code it generated when the User is from the USA / foreign? Might be good not to have that outcome again

Hey guys, I'm leading LLM research for fundamental research labs. I'll be posting more.

@Marvinma12 @Alibaba_Qwen Deepseek is going to want a piece of this action.

@Alibaba_Qwen The release of this model is more like China vs China in open-source, no other country comes close, now GLM will be next after Kimi and Qwen releases in a single week.

@vallcodes @Alibaba_Qwen Oh it’s been going strong for about 30 years now.

@Alibaba_Qwen OH GOD, THE OPEN SOURCE ERA IS COMING

Hermes Agent is my ride-or-die.

@XBusiness For the lizard people in your life.

Launch in minutes. Find customers for a lifetime. Try new X Ads Manager today.

bro disappeared like never existed

ScaleKit tested MCP against command-line interfaces across 75 benchmark runs. MCP consumed 4–32× more tokens. Agent Skills offer a better standard. groktop.us/stop-asking-fo…

@claudeai Does this mean subscriptions are discounted 50% until you figure out your logistics issues?

Beginning July 20, Claude Fable 5 will be included in all Max and Team Premium plans, at 50% of limits. Pro and Team Standard users will continue to have access to Fable via usage credits, and will receive a one-time $100 credit. Demand for Fable has been challenging to predict, which is why we rolled it out to subscription plans in stages, extending access several times as we secured additional capacity.

Don't let anyone tell you there's only one right way for work to flow through an AI agent. You've got many good tools at your disposal.

@tonysimons_ I've got a few things in GitHub that I'm proud of. But I think my agent skills are the true daily go-to. And every time I run into something I want to do with Hermes, I make a new skill for it and usually share it. github.com/magnus919/agen…

Hey Hermes Agent builders! 🪽 Show me your current pet project. I want to see the absolute best you’ve got and help you get more eyes on them. I’ll pick a couple from the comments and do deep-dive reviews this weekend. 🤘🏻

@insachins @steipete @grok Loops, graphs, and agents aren’t three generations of AI. Loops and graphs describe how work moves. Agency describes who chooses the next step. Workflows and agents can use all three. The simplest reliable system gets actual work done.

@steipete Loops were for automation. Graphs are for reasoning. Agentic workflows are for getting actual work done. @grok Can you comment on my thinking?

Are we still talking loops or did we shift to graphs yet?

@thsottiaux I just wish it wasn't compacting all the time in my agent. 272K context limit is really constraining.

GPT-5.6 Sol confirmed to be an extremely good model

@thsottiaux You can set this as a recurring task in ChatGPT now. Just sayin'.

Oops... I did it again. Enjoy reset usage limits for all paid users for Codex and ChatGPT Work. Super grateful for an incredible team who is iterating at lightspeed and keeping the infra up as we scale faster than ever. Enjoy the weekend!

So wait is #Exit8 just the #Backrooms, entered through a null zone in Japan?

@app_settings It's a useless feature. I don't want it. I saw it and I still don't want it.

why is EVERYONE in the comments all like “noo it sucks i want real function keys” omg DID YOU EVEN LOOK AT THE CONCEPT BEFORE REPLYING it has a Touch Bar AND a function row COME ON YALL Touch Bar hate is soo forced

Hermes Agent just needs a Backrooms skill...

Fable is gone. The great sucking sound you're now hearing is all of the Claude customers canceling their subscriptions.

@ComicBook_Movie Egghead confirmed!

Rumor: Scarlett Johansson's The Batman 2 Role Has Been Revealed - And It's Not A Character Anyone Expected