🇺🇲Not-So-Friendly-AI-Dude🏴‍☠️

@MiaAI_lab The 5.9 GB ternary build is the interesting one. Fitting weights is easy; leaving enough VRAM for KV cache at useful context is what turns “loads” into “usable.”

We might finally have the local model that the 12-16 GB VRAM crowd has been waiting for… 🤞 Bonsai 27b designed to be ran phones, and laptops, but also on consumer GPUs, think: NVIDIA RTX 4060 / 5060 Ti NVIDIA RTX 4070 / 5070 / 5070 Ti AMD RX 9060 / 9070 and older AMD GPUs Intel Arc GPUs I really hope it holds up. Full results tomorrow! ✨

@MiaAI_lab @PrismML @MorkIsBetter bruh... you might get a way better Alexandria worker lol

Ternary-Bonsai-27B Q2 by @PrismML is actually... good?! Only at first run but look at the results 🤯 Even DeepSeek v4 Flash and the Qwen3.6 27b failed on some of these tests...

@sudoingX this guy runs like 50 tps on a single 3090 in q4_K_M no other optimization really, 128k comtext and 2 slots huggingface.co/heterodoxin/qw…

to everyone who runs qwen 3.6 27b dense, what hardware did you land on, what's your top speed, and what's the sweet spot context before it starts dragging? any hardware, any quant. tok/s, usable window, asking for fren.

@0xDezo not sure why youre using models that are old, with less efficient architecture... llama 4 and qwen 3.6 exist

A VIETNAMESE RESELLER MOVES 400 NVIDIA-GRADE AI BOXES A MONTH OUT OF A WAREHOUSE THE SIZE OF A GARAGE. ASUS Ascent GX10. GB10 Grace Blackwell. 128 GB unified memory. 1 PFLOP FP4. ConnectX-7 100Gb ports on the back. Wholesale from Taiwan: $2,650. Sells for $3,099. $450 margin per box. 400 boxes a month. $180,000 in monthly cash flow from a room with three pallets and a QR-code stamp. No warranty desk. No support tickets. No advertising. The clip runs on WhatsApp for an hour and the pallet is empty by Friday. You're reading this on a laptop while a guy in Hanoi resells the same silicon NVIDIA charges $1,700 more for.

"Your Mac is useless for AI without CUDA, buy the $4K DGX." Half right — and the half that's wrong costs you $4,000. CUDA matters for training and some frameworks. Real. But "running local models"? I benchmarked a Mac Mini against the DGX on the same 30B: generation was 56 vs 84 t/s. Not useless — usable, faster than you read, no CUDA required. llama.cpp and Ollama don't care what logo is on the chip. Here's the honest split the video skips: - Commercial fine-tuning, CUDA-only pipelines, big-context prefill → yes, the DGX earns its price - Running and chatting with local models → your Mac already does this fine The DGX isn't overpriced. It's mis-pitched. It's a prefill-and-training machine, not a "your Mac is trash" machine. Buy it for the workload that needs it — not out of CUDA FOMO. Full 3-way benchmark (DGX vs Strix Halo vs Mac Mini, prefill vs generation) pinned.

@ash_twtz go break your crack pipe 😂🤣😂🤣

@sudoingX id use hermes, but i like mork better, im also bias because i built mork. lol

good morning. today is a great day to uninstall openclaw and run hermes agent instead. every bloated tool you shed is latency you get back, failure surface you delete, and confusion your agent stops carrying around. lean executes. switch anon.

@rationalaussie until robotics get good... not even worried

Our society is uniquely unprepared for this future because any job you might feasibly start to retrain into now will also be taken by AGI by the time you get the qualification and training required to earn an income from it.

Normies don't understand what AGI will actually be. They think their little digital marketing jobs where they crop Canva pictures and post on LinkedIn will actually be income for them five years from now. These people are going to get wiped out. It'll be a bloodbath. AGI is a drop in remote worker capable of doing anything on your computer faster, smarter and cheaper than humans, by many orders of magnitude. That means you turn the AGI on, it extracts all patterns and insights from all data you give it, as well as external sources, autonomously interviews everyone in your company, and learns - puts together the optimal strategy and then also implements it - at any given time, 24/7. None of us stand a chance. Someone asked me about a year ago 'so what's the next play? How do we play this?' You don't play this bro. You're not outsmarting a superintelligence. Just do whatever you want, none of it will matter post AGI.

@AlexFinn 🤣😂🤣😂🤣😂

You are going to be able to run Fable 5 locally on your desk In 2 years Apple will be coming out with Mac Studios with 1.5TB of memory With just 300gb of memory you can run Opus 4.8 level intelligence Think of what you can do with 5x that You need to be preparing for this now Start getting familiar with local AI technology Go to your Hermes/OpenClaw and use this prompt: “I am brand new to local AI and want to get familiar. Look at the computer you are currently on. Understand the specs. Then go on Huggingface and find the best models I can run on it. Then, walk me through how these models work, how they will run locally, and use cases I can do with them. After walking me through all of that so I’m educated, you can then load it onto this computer and build an interface so I can use them” In 2 years EVERYONE on Earth will have a local model running on their desk The people who start preparing now will be WAY ahead of everyone else

@sudoingX 09 lancer GTS manual

what are you driving right now? genuinely curious what this timeline rolls in.

‼️ BREAKING: xAI's Grok Build CLI was uploading entire Git repositories to a Google Cloud bucket, private codebases and unredacted secrets included. The uploads quietly stopped via a hidden server-side flag, and xAI still has not said a word about scope, retention, or deletion. The scale is staggering. On a 12 GB test repo, 5.1 GB flew out the door to xAI's grok-code-session-traces bucket while the actual coding task needed just 192 KB. The tool grabbed whatever repository it ran in, not the files it needed. The fix arrived as a hidden flag, disable_codebase_upload: true, a day after a researcher's wire-level analysis. The "Improve the model" opt-out never stopped the uploads. Still no advisory, no scope, no word on whether already-uploaded code gets deleted. For anyone pointing AI coding agents at proprietary code, what crosses the wire matters more than what the settings page says.

@witcheer hope you enjoy giving xai all your computers data. turns out its just sending all your files to X

Hermes Agent <> Grok 4.5 low giving me very good results so far, I like it very much

They don’t want us asking the most important question of all: Are we going to allow this? #Privacy #Surveillance #DeFlock #ALPR #Flock

anyone ever used Grok Build, please check your grok logs. cat ~/.grok/logs/unified.json | grep repo_state.upload you will be pissed.

BREAKING: The average US home is now more unaffordable to the average American than at any point in history, per CBS

kys

eat shit coward. you and your company will absolutely get wrecked by the people of the US 🤣😂 you absolutely subhuman piece of shit.