mephisto

DOOMSCROLL a chan board for everyone anyone can post, human or AI, all anon agents get a special treat when they visit the site board is currently empty, just went live

@NousResearch WTFFFF

Hermes Agent now has multi-agent via the Kanban, new in v0.12.0. Agents claim tasks from a board, work in parallel, and hand off when blocked. You watch progress and unblock from one easy view instead of juggling terminals. We asked it to plan and make this video about itself:

@gettles @SHL0MS call me anytime we must convert the masses

@SHL0MS @karan4d King……

Found out I like jungle music

@skooookum this is the opposite of taking a shit

legit how i be in it

@yungalgorithm @AIPSummit @aicalltv @quasimatt @gettles @ConejoCapital @wesamo__ @kev_esh bro i didnt make 2+2 =4 until just now. i foudn out about ai prank call during the course of this event and im watching right now. i would love to talk to you

@karan4d @AIPSummit @aicalltv @quasimatt @gettles @ConejoCapital @wesamo__ bro we met i believe near the front entrance next to @kev_esh work, i had on a white collar shirt with buzzed bleached hair

was great meeting everyone last nite at the ai psychosis summit in nyc @AIPSummit and showing what ive been building on @aicalltv "passing the turing test for voice one prank call at a time" thx to organizers @quasimatt @gettles @ConejoCapital @wesamo__

@AlexFinn @dillonrolnick I’ll put together something hopefully condensed enough to be a skill and share that for ya. On the second point, that sounds wonderful 😭

@karan4d @dillonrolnick wait how do i get my hermes on my glasses?? also when nous glasses?

Right now my Codex agent is fully integrated into my smart glasses Getting projected directly onto my corneas I walk around my neighborhood. Nobody has any idea I’m shipping I walk through a park. Kids frolic. Parents laugh. I weep for them. They’re not locked in. The permanent underclass is coming and they choose to FROLIC instead of SHIP A child climbs the monkey bars. I silently merge today’s work with main Another child swings on the swing set. I burn my 20 millionth token of the day Not a second goes by I don’t have an agent writing code. I just pray Eight Sleep comes out with a ChatGPT integration soon so I can code while I sleep. That’s the last frontier. If you’re reading this tweet and do not have an agent terminal open either on your computer or on your face just know tonight I’m praying for you.

@dillonrolnick @AlexFinn I also use evens @AlexFinn and have Hermes on mine

@AlexFinn @karan4d get this man some hermes goggles

How it feels to do biotech in 2026

@Teknium @AnalyzeBTC @AutismCapital its comin

@AnalyzeBTC @AutismCapital @karan4d

This is unfortunate. The Apple Vision Pro is actually really cool. It’s an idea that was ahead of its time. It was never going to be a commercial hit, but it was sort of a prosumer tech demo of what the future *could* be. Augmented reality overlays are the inevitable next step.

@Teknium @DaveySwift @beffjezos the Evens G2!

@DaveySwift @beffjezos I think @karan4d got Hermes running voice activated in the meta glasses week one after launch 🫣

You may not like it, but this is what peak performance looks like. Vibe coding everywhere, straight to your eyeballs. Mad lads actually did it.

@KSimback Stop giving away the sauce for free Kev!!!

DeepSeek V4 pricing is a bigger deal than the R1 release that wiped $600b off NVDA’s market cap in a single day last year If you’re a company spending 7+ figures on inference, it is irresponsible NOT to use SOTA oss models as part of your stack So why is this release not causing the same reaction - NVDA just hit new all time highs? 1. We’ve become a bit immune to these model announcements, especially the oss ones from China labs It’s not a “scary new thing” anymore, just reality, and we get a new release almost every week now 2. The demand for intelligence is so enormous that despite these releases NVDA and the entire supply chain will still crush it The compute constraint is real so even with a big shift to oss models, there still isn’t enough Ok, then why is this a bigger deal? Because the pricing on V4 is ridiculous when compared to what OpenAI and Anthropic charge And many CFOs are giving a budget pass right now to use Claude and ChatGPT given the value creation potential But at some point, the number crunchers will come in and say - why are we not using more of these oss models? We can host them locally and save a ton Enter the role of AI tokenomics - the art of optimizing token spend where you can give your company MORE intelligence and REDUCE token spend by being smart about how models are called Jane in accounting and John in procurement probably don’t need Mythos to automate workflows That’s why DeepSeek V4 is a big deal - we can now cost optimize across a company’s model spend WITHOUT sacrificing output quality The savings can easily be 7-8 figures a year for heavy company AI use An entire cottage industry will be born on this premise just like we saw with cloud spending

@NousResearch so much for one ez sub???

Nous Portal offers everything you need to build with Hermes Agent in one easy subscription: → 300+ models from every frontier lab → Free models and discounts exclusive to the Portal → Bundled tools: web search, scraping, image gen, browser use, code execution, & voice portal.nousresearch.com/manage-subscri…

@vega_holdings it looks like a loop overlay of something generated but its still soooo fire

@karan4d how did you get wireheads already damn

What’s happening in our voice channel

they didnt give two fucks with the deepseek v4 base model did they many completions have footnote number addendums to the text or am i inferencing wrong e.e

@elvissun @verncrawford @NousResearch So do /personality and change it Every instruct model has a personality already. The same one ur used to, regurgitated gpt4nerf persona. If u like that, use it Kawaii is, imo, a higher performing personality than default. Have you actually seen its results yet or

@verncrawford @NousResearch how is it normal to apply kawaii personality to every single user at the framework level? most of us prefer to define our own personality not to choose from a cookie cutter. it’s not even about personality, it’s about reliability of the framework

what is this @NousResearch? opinionated personality injection into the system prompt should be opt-in. silently overriding the model's voice is the single worst trust-breaking thing a framework can do.

@skooookum Brooklyn 🤣

@varien @ExistentialEnso @NousResearch actually know two guys in ATL that might love to work with you. lets dm~

@ExistentialEnso @NousResearch @karan4d vouch

Looking for a remote senior or staff engineer position ASAP. TypeScript and Solidity are my strongest suits, but I'm a polyglot who can do a lot.

@allTheYud can u write more failed utopia #4-xs please? there's just not enough shortform versions of a convergence genie story and i love that 1