yonder

@blacknredtext subnautica

@MikeisIRL @coah80 this is the poor option, rich kids back in the day paid for private server hosting just to host three people

@coah80 I wouldn't know this im not a rich kid

Hamachi was genuinely terrible nobody should ever defend this stupid software

@the_andrey_x you can press the power button 5 times to disable face ID and force PIN

The comments are full of people saying biometrics are the best... BIOMETRICS ARE THE WORST SECURITY SYSTEM Cops and soldiers CAN and WILL: 1) take your phone away and open it by holding it in front of your face 2) forcibly place your finger on your fingerprint ID There is no reliable way to prevent this from happening. Never use biometrics if you're an activist or working in a high-risk environment. INSTEAD: 1) Set a long alphanumeric password, change it periodically 2) On iPhones put the setting that erases your phone after 10 incorrect password guesses 3) Make sure your apps don't show your messages when your phone is locked 4) Set all your chats on auto-delete after a week 5) Set additional internal passwords on all important apps 6) If you're getting arrested/kidnapped, try to turn your phone off before it's taken away from you, on latest models this automatically encrypts it

@Aella_Girl huh make sense

a great test of "would you have been chill with owning slaves, if you'd been born into a slave-owning family" is whether or not you're vegetarian today

@lauriewired trvke

@kayleecodez hate to say it, but everyone that rejects kubernetes inevitably ends up recreating it from first principles lol

recently worked on an autoscaling system on top of morph cloud for the past few days and ngl it was hella interesting and some of you might not know what autoscaling is and what morph cloud is so lemme tell you basically autoscaling means your system automatically creates or removes machines depending on traffic/load. if your app suddenly gets a spike in users, more machines spin up. if traffic drops, extra machines get removed so you dont waste resources. sounds simple until you actually try building it. morph cloud basically lets you spin up cloud machines programmatically through APIs, snapshots, metadata etc, so my autoscaler could literally create and terminate VMs on its own depending on what the controller decided. the fun part is the architecture behind it. one loop in the system continuously observes reality by checking instances, health, metrics and provider state. another loop decides what the "desired state" should be. then both loops constantly try to reconcile with each other while the cloud provider itself is asynchronously doing things in the background. my fav bug was when the controller started creating machines faster than the collector could detect them, so the autoscaler was like "damn still not enough servers" and kept spawning more morph VMs every few seconds cuz of stale state and eventual consistency issues. also learnt that in infra software, false success is genuinely scarier than failure. if the cloud provider rejects an instance creation request but your control loop records it as success anyway, your entire system starts believing imaginary machines exist and everything slowly turns into some fuckery. honestly one of the coolest things ive worked on recently.

@dudeman6790 @loktar00 wow it was $3000 when i bought it

@loktar00 This one does amazon.com/ASUS-Ascent-GX…

lol wut? Amazon has the DGX Spark just listed as "High price" 😂

@varrock "diversity" according to leftoids is the act of mixing everything together into brown, thereby completely destroying actual diversity

@sudoingX if qwen 3.6 122B launches, this will be the best model for single spark!

for the last few weeks i kept dodging the "is the dgx spark worth it?" question because i did not have one to test. now i do, and i can give you the real answer. it is an absolute workhorse. cranks 56 tok/s on a 30b model at q8 with full multimodal + tool calls on hermes agent, eats long prompts at 1,300 tok/s prefill, holds 256k context without breaking a sweat. 128gb unified memory unlocks model classes nothing else in the consumer tier even tries. low maintenance, silent, sits on the desk and works. if you can afford it and you are serious about local ai at the frontier consumer tier, this deserves a slot on your desk. one of the rare pieces of hardware that earns its price the moment it powers on

@runsonai @lud_ahr @sudoingX it's $99 fs.com/products/11563…

@lud_ahr @sudoingX Just looked up the price of the cable…$1500?! Crazy.

@marktilbury @CryptoScoopOG

I'm British. The UK is a warning to the rest of the world. Please listen very carefully:

@ImGregPartlow @yacineMTB "massive" here is hundreds of megabytes, it's practically nothing

@yacineMTB static linking simplifies deployment but binary sizes get massive fast

Everything should be statically linked. Dynamic linking is idiotic

@ssskryl @yacineMTB so much better than having to deal with dependency issues

@yacineMTB I love it when my CLI tools are 500mb and come bundled with a fucking browser runtime.

@jonasfroeller @moo9000 looks like gitlab enterprise 😂

What if the EU built GitHub?

@TeksEdge the same system on different cases have been around for months lol, most likely cheaper than what AMD will be selling these for

⁉️So get this, AMD is making a bold move to own the affordable personal inferencing market by launching a Mini PC in June, a 128GB Shared Memory Inferencing Box 🎇 They call it the ⬭ Halo Box. 🧾 It's a Ryzen AI MAX+ 395 (16 Zen 5 cores + 40 RDNA 3.5 CUs + XDNA 2 NPU) ✅ Up to 128GB LPDDR5X-8533 unified memory ✅ Full ROCm support + Day-0 AI model optimization 🧪 Built for local AI development (up to ~200B param models) 📈 Direct shot at NVIDIA’s $4,699 DGX Spark and could cost $2,000–$3,000 (as they do now) 🤔 Why launch now during the RAM shortage? While memory makers divert capacity to HBM for AI data centers (driving LPDDR5X prices to spike and NVIDIA to raise the price of DGX Spark by $700), AMD is making a bold move to own the affordable, high-memory AI mini-PC segment before the crisis worsens. 💡 My Speculation: AMD could be using its contracts, relationships, and strategic priority to secure better memory access than many traditional OEMs. This could give them an advantage in launching the Halo Box during the shortage. Smart timing or risky bet? 🔥 This is AMD aggressively fighting for the local AI developer market.

@alexckshen @ArtificialAnlys @xai if 4.3 really is 500B then that is actually extremely impressive

@ArtificialAnlys @xai Why tf is *looks again* xiaomi ahead of xai 😭

xAI has launched Grok 4.3, achieving 53 on the Artificial Analysis Intelligence Index with improved agentic performance, ~40% lower input price, and ~60% lower output price than Grok 4.20 The release of Grok 4.3 places @xAI just above Muse Spark and Claude Sonnet 4.6 on the Intelligence Index, and a 4 points ahead of the latest version of Grok 4.20. Grok 4.3 improves its Artificial Analysis Intelligence Index score while reducing cost to run the benchmark suite. Key Takeaways: ➤ Grok 4.3 improves on cost-per-intelligence relative to Grok 4.20 0309 v2: it scores higher on the Intelligence Index while costing less to run the full benchmark suite. Grok 4.3 costs $395 to run the Artificial Analysis Intelligence Index, around 20% lower than Grok 4.20 0309 v2, despite using more output tokens. This makes it one of the lower-cost models at its intelligence level ➤ Large increase in real world agentic task performance: The largest single benchmark improvement is on GDPval-AA, where Grok 4.3 scores an ELO of 1500, up 321 points from Grok 4.20 0309 v2’s score of 1179 Grok 4.3, surpassing Gemini 3.1 Pro Preview, Muse Spark, Gpt-5.4 mini (xhigh), and Kimi K2.5. Grok 4.3 narrows the gap to the leading model on GDPval-AA, but still trails GPT-5.5 (xhigh) by 276 Elo points, with an expected win rate of ~17% against GPT-5.5 (xhigh) under the standard Elo formula ➤ Grok 4.3’s performs strongly on instruction following and agentic customer support tasks. It gains 5 points on 𝜏²-Bench Telecom to reach 98%, in line with GLM-5.1. Grok 4.3 maintains an 81% IFBench score from Grok 4.20 0309 v2 ➤ Gains 8 points on AA-Omniscience Accuracy, but at the cost of lower AA-Omniscience Non-Hallucination Rate of 8 points, so Grok 4.20 0309 v2 still leads AA-Omniscience Non-Hallucination Rate, followed by MiMo-V2.5-Pro, in line with Grok 4.3 Congratulations to @xAI and @elonmusk on the impressive release!

@obrien chinese models are better in every way possible especially cost lol

I get people dunking on Mistral, but if you are in Europe looking at this: does your agentic deployment need frontier capability of Claude, 'cause that inference bill will be insane. so, then you pick a Chinese model that may or may not be marginally better in some verticals?

@scaling01 weaker than 300B MoEs..

Mistral Medium 3.5 is out and it's a dense 128B model

@ThiccQuidity @Hedgeye they're for living, not speculation. other financial instruments are better for speculation

@Hedgeye imagine buying a house, and 20 years later it’s worth as much as the day you bought it

🚨 China's Real Estate Market has erased all gains from the last 20 years

@sama prove it by releasing a new OSS model chud

we love our users

@XiaomiMiMo woah!! 310B!! thank you!!

Xiaomi MiMo-V2.5 is now officially open-sourced！ MIT License, supporting commercial deployment, continued training, and fine-tuning - no additional authorization required. Two models, both supporting a 1M-token context window : • MiMo-V2.5-Pro: built for complex agent and coding tasks, ranking No.1 among open-source models on GDPVal-AA and ClawEval • MiMo-V2.5: a native omni-modal model with strong agent capabilities A model's value isn't measured by rankings alone — it's measured by the problems it solves. Let's build with MiMo now! 🤗 Weights: huggingface.co/collections/Xi… 📄 Blog: #blog" target="_blank" rel="nofollow noopener">mimo.xiaomi.com/index#blog