Rob_801

@morganlinton I have 2 home pod minis, and tbh I rarely use them. They’re not “big” enough. More often than not, I play through one of my Samsung soundbars. Really disappointing honestly because I’d rather have an amazing product from Apple because for use, the HomePod wins hands down.

Remember when everyone said this was going to be the Sonos killer? Cheaper than Sonos, made by Apple, everyone said Sonos didn't stand a chance. How many people reading this have ever gone to a house that plays music throughout the house with a HomePod? Now how many with Sonos. Just because something is cheaper, and from a bigger company, doesn't mean it's better. And I love Apple, but Sonos does just one thing, Apple does a lot of things. I feel the same way about Google with the FitBit Air. Everyone seems to be calling it the WHOOP killer. Focus always wins.

@Renegade_Rocks @atomicbot_ai Just run Hermes migrate. Tell your Hermes agent what you’re doing, what you need and what you want and you’ll get amazing results.

@801Rob @atomicbot_ai As much as I am a fan of Openclaw, I have to agree hermes always comes back with the job done, no hiccups, no strange behaviours. I think about migrating to hermes daily but just been holding on as I am worried about all my field in the Openclaw workspace. Will they carry over?

Hermes Agent vs OpenClaw using Qwen 35B Local Model We asked agents to scrape GitHub star history for both tools, find what caused the growth spikes, build a live dashboard in the browser. MacBook Pro M5 Max 64Gb OpenClaw: 203k tokens, 12m 01s - wrote a bash script Hermes: 257k tokens, 33m 01s - wrote a SKILL.md OpenClaw hit GitHub API, got truncated responses, paginated through contributors, pulled star-history JSON, found a security incident in OpenClaw's history, fetched SVGs, fixed broken HTML from trimming, rewrote it clean. Hermes parallel tool calls across GitHub API, web search, and browser. Hit Google rate limit, auto-switched to DuckDuckGo. Fetched article contents, mapped viral moments, then built the dashboard. Both shipped a live dashboard with star growth charts and spike annotations

@ivanfioravanti I was not aware you could run oAuth with grok? Interesting..

To get the most out of Grok 4.3 + Hermes Agent configure it for text and in all tools for images, videos, text-to-speech and X search! This + 1M context... BOOM! 🚀

For hardware, Mac mini. Main model now is GPT 5.5 via codex on my plus sub. It doesn’t matter what model I’ve ran though, I’ve ran 30+ models. I’ve ran Kimi, Mimo, Nemotron, Gemini, Qwen and idk how many others. I’ve ran local, cloud via openrouter, Claude via API, it literally doesn’t matter. Every model I’ve ran has had almost equally great results. It’s the harness (ie Hermes itself), not the model. @Teknium made something TRULY special with Hermes.

@801Rob 3 weeks no downtime says it all honestly. what are you running on it though?

@0x_aut @0xNdrej @0xSero You’re literally starting shit, to start shit. If you don’t like it, move the fuck on then. You’re digging for something you ain’t gonna find.

@0xNdrej @0xSero However a problem is the quality of what he releases, I don’t claim to know much but if it can’t pass a certain quality line for me, I wouldn’t like it

Like someone without anything useful to do, I will now attempt to go through the @0xSero logs and decide for myself whether he is legit or not, there are also talks of him being a crypto bro in the past? Oof

@morganlinton That’s, insane.. I have some ideas I could do with this, but I’d hate to spend the time/tokens and not be happy with the result.

Back in early May, I decided to put Codex to the test, to see if, with a single prompt, it could build an operating system from scratch. It's still going, over five days of total time spent pursuing the goal. While I can't run all day every day, doing this all within my plan limits which are 10x thanks to the generosity of @sama and team for going to the GPT 5.5 event 🙏

@MartaCecilia36 Where you’ve got it wrong is “checkout as guest” because I’m not signing up for shit. Order Domino’s on my way home, it’s ready in 12-15 minutes, and I go to pick up my own damn pizza like a true Gen-X savage and hand deliver that shit to my family.

Pedir una pizza a domicilio 1995 – Llamas por teléfono – La pides en 2 minutos – Llega en 30 minutos 2005 : – Entras a la web – La personalizas – Llega en 45 minutos 2026↓↓

@vmiss33 I’ll be setting up one for each of my three sons on their computers as soon as it’s ready to go.

Hermes Agent on Windows. This is a huge deal to get it into more hands.

@ivanfioravanti I’m literally the bottleneck. I can’t keep up with my agents, not even close.

Full-time on AI and I can barely scratch the surface of the amazing things to test, try, and build. We're living through the most incredible era and it keeps accelerating! 🚀

@3PeaksTrading Do you post these trades in Optionshawk?

Pretty wild how much implied vols are expanding under the surface past few days... can see it just in my $SPX diagonal spreads lol big move there so may have to take some profits but better to be a vol buyer here imo

@witcheer @NousResearch Hermes is literally GOAT. I think I’m now ~6 weeks uptime without a single failure or issue. It just works, like every time for every thing.

.@NousResearch agent is now #1 globally on openrouter. 6.08 trillion tokens processed. ahead of openclaw, kilo code, and claude code. I’ve been running hermes on a mac mini M4 since february. daily peg monitoring, research crons, draft reviews, wiki maintenance. it felt like a niche tool for people willing to configure things themselves. 6 trillion tokens says it isn’t niche anymore. 6,600+ builders in the community. fastest-growing AI agent ecosystem on X right now. what makes hermes different from the rest of the list is that it’s fully open-source, self-improving (skills created from experience, auto-curated on a 7-day cycle), works with any model provider (200+ via openrouter, or local via ollama/vllm), and runs on a $5/month VPS or a mac mini on your desk. Hermes for the W

I’ll keep commenting on every, single, fucking, one of these posts. HERMES, just get Hermes, run Hermes migrate. My agent has been up for ~3 weeks CONTINUOUS, I’ve had zero issues, countless updates, multiple projects done, it just fucking works. Run GPT-5.5 as your main, a cheap fallback on openrouter or cheaper GPT and you’ll never look back. I even have Grok api for twitter stuff, it’s literally amazing having an agent that works and you can’t comprehend how good it is until you just do it. There’s a reason Hermes is having such explosive growth, it, just, fucking, works..

I don't know if this is a skill issue or something... But I've been playing with OpenClaw a ton for the past 4ish months. I feel like lately I'm spending more time troubleshooting issues with it and telling it what it's doing wrong than I am actually getting valuable use from it.

@rileybrown The e easy fix to this is Hermes migrate

OpenClaw starts out very fun, then the headaches start happening after a few weeks. I’ve talked to many AI agent power users who are also struggling with OpenClaw. Gateway issues, cronjobs not firing, skills not being used consistently, memory fall off, integrations need re-authentication, oscillating between overeager and undereager, not placing files in the right place. Etc

@NVIDIA_AI_PC Hmmm, like 6? I need to go download all the NVDA ones though, just been busy.

Be honest — how many local models do you have downloaded right now? 👀

@Teknium Thinks absolutely incredible, but I have absolutely no idea what I would actually do with it.. :/

👀👀

@bradmillscan Your fix is easy, it’s called Hermes migrate.

the last 2 version of OpenClaw are not working for me. Agent chat has degraded significantly, and I finally found the culprit. context overflow happens because of excessive tool use, and then openclaw re-injects the last message you sent back to the agent. That's why the agent responds to the same thing over and over. I was having this excessive tool use problem back in March when I was last heavily using OpenClaw ... but it was giving me noisy errors ... now it just silently degrades and fails. The root cause of the problem didn't get fixed ... excessive tool use polluting the session and causing context overflow. The failure mode just changed from a warning message to off-by-1 chat degradation.

@cryptosibbers @NousResearch Bro, you’ve only glossed over the 800 other reasons to switch if this was the one that made you jump.. lol

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:

@trkbt10 I cancelled my sub yesterday and abandoned 2 projects, the burn out from all the lying and mistakes from Claude just became too much.

Opusこれ今本当にひどい有様で、なんでこんなんなったのか説明してほしい かなり本当に酷い

I downgraded to the $20 Claude plan today. Claude degraded so badly, so fast that it was no longer worth it. I get more value, better results, and FAR more value using chat gpt/codex and droid along with Hermes for $80 Less a month. The second M5 Ultra Studios are out, I’ll have local AI for life.

Most logical video on what will happen to our AI subs soon. youtu.be/q5co-Ptjz0o?is…