John B. Manos

@tonysimons_ I hope it doesn’t bring the enshitification common with VC funding.

I’ve been telling y’all Hermes Agent is different. Now Nous Research is reportedly raising at a $1.5 BILLION valuation. At least $75 million in new funding. For the open-source AI lab behind Hermes Agent. Yeah... people are about to start paying a lot more attention. Here's a full breakdown of what this could mean for Nous, Hermes, and the open-source AI race: ⤵️ tonyreviewsthings.com/nous-research-…

@mwmoedinger Surely they’ll “improve” the service by stripping away all that is likeable with the current and making the UI more and more obtuse as performative gestures of value add.

One of my cornerstone, much-beloved software was acquired last year, and yep, you guessed it...they just increased my monthly plan from $22 to $248. It's such BS, bc now I get to spend a ton of time looking at alternates, migrating 12 years of data, changing a bunch of internal processes, and wasting a sh!tton of time all bc some asshat PE company did the math on how many customers they can lose vs how many will be too tired to fight the increase. Please *do not* suggest any alternates. *Do not* suggest I can build something in Claude. That is not the point. If you own a small business, you understand. If it's not one thing, it's another. The grind is....relentless. (Related: now accepting cookie and amaro deliveries at my office.)

@jessegenet I don’t know if bringing the gaslights and propaganda into your homeschool is what you really want

Great! Now do verified homeschooling parents 🤓

@tonysimons_ Same here! LOL

The long and short answer to this question: Hermes Agent 🪽 I may not have a clue what I’m doing. I admit that. But I’m still doing it. Things I’d never be able to do if it weren’t for Hermes.

@MiaAI_lab Im still wondering why no one complains about Claude retaining all your data for 30 days (even if you tell them not tontrain on it).

The only correct move.

Wordle 1,851 3/6 ⬛⬛🟩🟩🟨 🟨⬛🟩🟩⬛ 🟩🟩🟩🟩🟩

@techdevnotes She had no weekly limits … err no limits at all from what the buzz was like then

1 year ago xAI released Ani

@appleinsider @MalcolmOwen Hmmm. Nothing like the m4 ultra ?

It takes years to design and fabricate a chip, with overall industry seismic shifts taking much less time. Nothing will demonstrate that better than the rumored six-month gap between the M6 and M7. By @MalcolmOwen appleinsider.com/articles/26/07…

@saen_dev @Hesamation Not accidental. Been the plan since Taligent.

@Hesamation 1.5TB unified memory means running full unquantized 400B+ models locally on a single machine. Apple isn't making consumer hardware anymore, they're accidentally building the cheapest inference servers on the planet. What's the estimated memory bandwidth on this chip?

Apple’s M7 Ultra chip rumored to be a local AI workhorse as it supports 1.5TB of RAM. for reference, that’s enough to run Kimi or GLM models locally.

@kai_benetti @Hesamation We already have that

@Hesamation We are basically going to have full data centers on our desks soon

Wordle 1,850 5/6 ⬛⬛⬛⬛🟩 🟩⬛🟩⬛🟩 🟩⬛🟩🟩🟩 🟩⬛🟩🟩🟩 🟩🟩🟩🟩🟩

@MacRumors Checks notes. Yeah i don’t see an M4 ultra anywhere…. Guess the ultra skips even

Here's Why Apple is Reportedly Skipping M6 Pro and M6 Max Chips macrumors.com/2026/07/12/why…

@sudoingX Now that is kind of tricky for me because I was running expert systems in the 90s

what was the first gpu you ever ran a local model on, and what are you on now? give me the first card and the current card, and the year gap if you remember it. i want the ladder, not the flex.

@Teknium Oh man! I lost my 17 yr old recently and still miss him daily!!!

Hey everyone. I haven't been very responsive on here the last week. My dog, Link, who I've raised since he was a puppy over the last 13 years, passed away yesterday after being in the vet ER's ICU since last Wednesday for heart failure. I put together some of my favorite pics of him to share so you all can see the most awesome animal friend I could ask for. I'll be a bit slow probably through this week too, hope you all can understand 🙏

@LuminaXspace Funny, I felt that supergrok was dang near unlimited for the past year and a half - all the way up until they sold out to Wall Street.

🚨 Grok’s usage limits are genuinely awful: • I’ve barely used SuperGrok this week and somehow I’m already at 65%, chat alone using 49%, that is using skills and expert mode which chunks 1-2% per prompt. • Decided to try out ChatGPT and it feels almost unlimited, I can't hit a single limit even spamming codex. • SpaceXAI seriously needs to improve its paid-plan limits, I love using Grok but im not paying for it anymore, I have to limit my use just so I don't run out or get 2 days out of the week to actually try it using it properly. Have you felt the terrible usage limits and have any of you unsubscribed yet?

@SteveOnSpeed How bout twice a day like a real first world

Here’s an idea: Deliver mail 2 to 3 days a week instead of 6. All most people need.

@brennen_harris_ @carlos_med82 @elonmusk @SpaceXAI @grok Actually, I’ve had the $30 subscription for a year and a half. I know what to expect and it was way higher until recently.

@carlos_med82 @elonmusk @SpaceXAI @grok Can’t expect too much out of a $30/month sub. High usage volumes on a frontier tool that’s used for work shouldn’t cost as much as your Netflix subscription If you have real world application and use cases then snag a heavy sub and you’ll have plenty of usage available.

The usage limits on the $30 SuperGrok subscription are much lower than they used to be. Please fix this or I’ll have to cancel my annual subscription.@elonmusk @SpaceXAI @grok

Wordle 1,849 5/6 🟩🟩⬛🟨⬛ 🟩🟩🟩⬛⬛ 🟩🟩🟩⬛⬛ 🟩🟩🟩⬛🟩 🟩🟩🟩🟩🟩

@drstrangelau @xai @grok @imagine Wall Street got involved.

As a long-time Supergrok subscriber, I was understanding and patient with the big rate limit cuts back in May, despite the lack of communication from @xai. Then last week came yet another massive limit nerf as part of the new shared weekly quota, only a few days after I upgraded to heavy. It’s brutal. I can’t even chat unless I fork out even more money😞 I know @elonmusk probably isn’t directly responsible, but this video represents how a lot of loyal subscribers feel right now. Made with Grok Imagine @grok @imagine

@jdkiker_X Wall Street in the house sucking value out of the room. Amazing how fast they did it, too

SuperGrok is super useless now. Every feature, that is not basic chat, makes it less usable with each request. Definitely will not be putting more money into this product.