JASON MCNAB

Codex and cheer for grok to become real player soon. Anthropic failed to plan to scale. And has made decisions for the fail to scale that alienates customers. Watch them work with government to make some AI “safe” so others are “unsafe.” Destroying opensource pipelines and securing their monopoly. They are desperately trying to increase liability that doesn’t exist with any other product. If a moron jumps a curb and runs over people, the manufacturer of that car isn’t liable. They are trying to do that to AI. And if you disagree, you are against the children. Anthropic will comply to be guaranteed a position in the market knowing it folded doors to competition behind them. This is a classic progress move littered throughout the early progressive movement in early 1900s.

I can't help but feel personally burned by the Claude Code changes announced today. We put so much work into wrapping the (atrocious) Claude Agent SDK in T3 Code. It was the ONLY path they supported, so we made it work. It was hell. Now our users are getting their rate limits cut by 40x, despite us doing everything right. I listened to the Claude Code team. I had my issues with their direction, but I trusted them and took them at their word. I will never make that mistake again. Until we see significant change, it is safe to assume any statement from an Anthropic employee is a lie on a timer. The rug will be pulled, no matter how many promises are made beforehand.

I try not to be a purist. It’s generally a mistake to be one. My preference is local. My control. That being said I can think of one use case that makes it worth it. If you want a Hermes agent to monitor and or assist you with your network when your network is experiencing issues. If your network at home is in trouble that outside help is an asset. Also to really pen test your network, it’s better to have your asset outside the network, correct? So there are use cases to have a Hermes or Openclaw agent outside your network on VPS. But local should be your primary.

I’ve been thinking about whether Hermes Agent makes more sense locally or on a VPS. Honestly? It depends. If you care about privacy, always-on workflows, or high-volume automation, local starts to look attractive. But for casual use, cloud is still simpler. Thoughts? 👇

@tonysimons_ @NeoAIForecast Both of you are awesome. In the ham radio community, they call guys like you elders. We are the newbs. I devour your guys posts and articles.

@NeoAIForecast Thank you, Neo! I appreciate you taking the time to say this, bro!

I don’t even know why I allow myself to go on Facebook these days. It’s such a cesspool of toxicity and a blatant reminder of how messed up some sectors of people in our society are. I’m very grateful for the community I have here on 𝕏 . I haven’t had one argument on this platform since starting this account in January. Such a pleasant change from the way OG Twitter used to be. If you’re seeing this and have engaged with my posts or are following me, thank you. I appreciate you all more than a tweet could ever express. 🫡

I been a lot happier and more tired (late night grinding) when I reset my X algorithm and made it all AI. They still try to get you in the "For you" with race gender class warfare bs. I just click the "not interested" in this post. I am over that fing game. It's a trap. It isn't real.

Don't stop grinding... you are doing awesome. Same for me. Thank God my GF is awesome and deals with it.

Always remember, In 1972, a crack commando unit was sent to prison by a military court for a crime they didn’t commit. These men promptly escaped from a maximum security stockade to the Los Angeles underground. Today, still wanted by the government, they survive as soldiers of fortune. If you have a problem, If no one else can help and if you can find them. Maybe you can hire, The A-Team.

@0xSero LOL after spending months begging me to create content with him he did the same thing to me yesterday. I'm pretty sure his whole MO is to just try to tear down anyone that's even mildly more successful than him He has to be the angriest human being on planet Earth

3 months ago I blocked this guy and he made such a scene about it that to this day people still ask me why. The reason I did so is because despite him hyping me up, he’d constantly be writing about how I’m a larper. Now my “larping” has resulted in: - Meeting folks at Nvidia - Meeting folks at OpenAI - Working with Factory - Teaching 1000s of people 100+ GitHub repos: - day 0 deepseek-v4-flash on sm120 - best performance on Framework - REAP-MLX + REAP-Strix - VLLM-STUDIO nearly 1k stars - GLM-4.6/4.7 on a MacBook - Qwen-3.5-plus for 8x 3090s - Parchi - AI-data-extraction 1k stars - First working turboquant on vLLM 4 months ago: - Interned at a large AI company - Produced 15+ models with 100k monthly downloads - Created a discord server and taught nearly 1000 people for free (still doing it) 6 months ago: - Released the first REAP quants - Sponsored by Anthropic running Claude code Warsaw with 500+ attendees - Trained nanochat at home 12 months ago: - Built my first rag self hosting on a MacBook funny enough - Spending 5-10k a month in tokens on random product buildings 18 months ago: - Taught a 200 person course (for free) how to use AI for coding 24 months ago: - Applied research for the Ethereum foundation on ZK proofs - Built Rosetta Node a solidity <> English translator built on OpenAI

I run Hermes Agents and Codex is when I sit down and vibe code. Codex is also my IT department. When there is a problem with Hermes or I need assistance perfecting a Hermes workspace to perform a task, I use Codex. Codex makes entries into my hive mind system with my Hermes Agents in the shared agents folder. So everyone has context. What this does for me, no more using apple app screens and expanding sections of a remote screen. No more painstakingly trying to get to a terminal through a shared screen and using the smallest clicks to be accurate on a Remote Desktop. I can get support remotely for my infrastructure at home. I love this. I can't wait to test it.

The DGX Spark community is incredible. Check out a DGX Spark here amzn.to/4dIVguB This is the best device IMO. Macs are gone and the wait is two months. These are available and won't be for long, trust me. I own 4 of them. Jump in and check out the forum here. forums.developer.nvidia.com/c/accelerated-…

@SamJWasserman @NVIDIAAI @nvidia @NVIDIAAIDev Thanks bro

@JASONMCNAB @NVIDIAAI @nvidia @NVIDIAAIDev I used Geekpi - great prices on amazon - 150 or under for this 12U - and great quality. Easy to build.

@NVIDIAAI @nvidia @NVIDIAAIDev Reposting my Nvidia Sparks, mac mini, 12u agentic rig is an epic nerd out moment for me. It's like if Jimmy Page (Zeppelin is my favorite band) reposted my music collection.

@PolicyProtest @dogeofficialceo Was he bastardizing a story for woke bullshit?

@dogeofficialceo So when Shakespeare had women characters played by men… you think that was woke?

Christopher Nolan directing The Odyssey

@aijoey Bro is this operating codex remote on your phone and codex on your gear at home? Being on the road for 3 days all day is like being out of what's happening for 6 months in AI.

Codex just won mobile. Way too smooth and fast.

This is amazing. Hermes and AI should be fun. It is work but can be fun too. It's not even fair to say this is just fun, this is amazing. I love this local AI hobbyist/professional community. Great work @outsource_

Looks like @NVIDIAAI has released official getting started guide for hermes on DGX Spark! Great way for new folks to get started and believe in local agentic AI

@AlexFinn @TheAhmadOsman Responsibility = liability

You can take any clip out of context. In context, when someone asked when starting out with no real business yet he advocated an old laptop or Mac mini. In the context of Mac studios and DGX it was someone with a business already and established workflows to hand off to openclaw agents and Hermes. We are part of a community of hobbyists. We should support each other. We are facing grotesque government regulation from all political sides that threaten the open source pipeline through manufactured fake responsibility. We need to approach this as a community with different opinions but strong commonalities in local hosting and individual sovereignty. We all have dignity. Let’s serve each other and grow this community.

This is Alex Finn He’s costed so many people their hard earned money during his Mac mini grift Now he’ll reuse the same script for the DGX Spark He doesn’t know how to highlight any hardware strengths/weaknesses Zero substance or knowledge of local AI Go grift something else

@HermesAgentTips Qwen 3.6 27b or 35b and Nemtoron OMNI as a collector 24/7. I found Gemma sliding context window annoying for long runs.

local LLM people: what are you actually running right now? everyone talks like they have a DGX Spark under the desk, but I’m curious what the real setups look like DGX Spark 128GB unified memory? RTX 6000? RTX 5090 32GB? RTX 3090 24GB? MacBook Pro? Mac Studio M3 Ultra, if you somehow found one? or are you running something completely different that people are sleeping on?

Love this. Good job Eric.