Nuno Sousa

Doing something different as a trial. I'm going to open majorbase.com up for other people to post their articles with their affiliate links. Rules: - No bad content - Content has to be about Gaming, Esports or Tech - Backlinks must be legit websites - No custom HTML

Open-source robotics ecosystem is exploding 🔥 Just look at the LeRobot ecosystem growth over the last 9 months: 📈 LeRobot Datasets: Up ~600% 🤖 LeRobot Models: Up ~4000% 😮‍💨 These numbers indicate just how hot Physical AI is nowadays.

@varun_mathur In what ways does this differ from @exolabs ? Genuine question. Is it the NAT traversal?

Introducing Pods Hyperspace Pods lets a small group of people - a family, a startup, a few friends, to pool their laptops and desktops into one AI cluster. Everyone installs the CLI, someone creates a pod, shares an invite link, and the machines form a mesh. Models like Qwen 3.5 32B or GLM-5 Turbo that need more memory than any single laptop has get automatically sharded across the group's devices - layers split proportionally, inference pipelined through the ring. From the outside it looks like one OpenAI-compatible API endpoint with a pk_* key that drops straight into your AI tools and products. No configuration beyond pasting the key and changing the base URL. A team of five paying for cloud AI burns $500–2,000 a month on API calls. The same team's existing machines can serve Qwen 3.5 (competitive on SWE-bench) and GLM-5 Turbo (#1 on BrowseComp for tool-calling and web research) for free - the hardware is already on their desks. When a query genuinely needs a frontier model nobody has locally, the pod falls back to cloud at wholesale rates from a shared treasury. But for the daily work - code reviews, refactors, research, drafting - local models handle it and nobody gets billed. And when it is idle, you can rent out your pod on the compute marketplace, with fine-grained permissions for access management. There's no central server involved in inference. Prompts go from your machine to your pod members' machines and back: all of this enabled by the fully peer-to-peer Hyperspace network. Pod state - who's a member, which API keys are valid, how much treasury is left - is replicated across members with consensus, so the whole thing works on a local network. Members behind home routers don't need port forwarding either. The practical setup for most pods is three models covering different jobs: Qwen 3.5 32B for code and reasoning, GLM-5 Turbo for browsing and research, Gemma 4 for fast lightweight tasks. All running on hardware you already own. Pods ship today in Hyperspace v5.19. Model sharding, API keys, treasury, and Raft coordinator are all live. What Makes This Different - No middleman. Your prompts travel from your IDE to your pod members' hardware and back. There is no server in between reading your data. - No vendor lock-in. Pod membership, API keys, and treasury are replicated across your own machines using Raft consensus. If the internet goes down, your local network keeps working. There is no database in someone else's cloud that your pod depends on. - Automatic sharding. You don't configure layer ranges or calculate VRAM budgets. Tell the pod which model you want. It figures out how to split it across whatever hardware is online. - Real NAT traversal. Your friend behind a home router with a dynamic IP? Works. No VPN, no Tailscale, no port forwarding. The nodes handle it. - Free when local. This is the part that matters most. Cloud AI bills scale with usage. Pod inference on local hardware scales with nothing. The marginal cost of your 10,000th prompt is the electricity your laptop was already using. Coming soon: - Pod federation: pods form alliances with other pods. - Marketplace: pods with spare capacity can sell inference to other pods.

@andreasklinger I think every European wants Europe and EU as a whole. What they don't want is that unelected dictators rule them and extort them of their money and resources to hand over to someone else that they had no saying on. Among other injustices that Brussels enforces on EU members.

Europe needs to pick. Be Latin America or North America. Details can be discussed. But its a fundamental question of our generation. How much Europe does Europe want. We need to decide that or others will decide for us.

Sneaky2x - Final Boss (Full Track Official Release) is now on my channel go listen 🎶 youtu.be/HJD9M1D7D1E?si…

Sneaky2x - Final Boss Reloaded (Full Track Official Release) is now on my channel go listen🎵 youtu.be/2K6G6lAbT4U

Sunrise Loot Squad is now here! 🎵 This one is dedicated to the teammates that keep you up at night until sunrise, gathering loot, picking fights, grinding and laughing. youtu.be/TxmSKgsnrxg

Raid Nights is now here! Go check it out, I hope you enjoy. 🎧 youtu.be/MdPNmbbVYuY

@vr_mama @TheRealQuenny @sean_gause Not true, there proper models out there and some Game Devs are using AI 3d model generators that can output 3d objects that can be rigged and animated just like any other proper 3d object But if you want texture extraction and eye rigging, tracking, that's a different talk

@Sneaky2x @TheRealQuenny @sean_gause Sadly thoae arent actual 3d models. Theres no visemes, eye tracking, data or ana actual model to use. Its an interactive image. You wouldnt be able to successfully post it to blender or Unity or even unreal.

Blocked me so I can't reply, but: - Everyone starts out as a "non-modeler" - Blender is free and open source - Blender doesn't spit out slop that's built on stolen data I wasn't born with 3D modeling skills, and yet I'm making the project of my dreams. No excuses.

@vincent_koc @NA8STRADAMUS @MatthewBerman OP is engagement baiting. This guy clearly doesn't know what he's doing. You guys are doing a great job, keep it up! 👏

@NA8STRADAMUS @MatthewBerman "support my app" which one? "my channel" which one? "depreciated", they were all backwards compatible with updates provided to alternatives

I have to say something...please don't be mad. OpenClaw has been nearly unusable for the past week. Something changed and now everything is broken. Looking forward to testing Personal Computer.

We’re open-sourcing HY-World 2.0, a multimodal world model that generates, reconstructs, and simulates interactive *3D worlds* from text, images, and videos. Outputs can be integrated into game engines and embodied simulation pipelines. Key highlights: 🔹 One-click world generation Turn text or image into interactive 3D worlds automatically. 🔹 Pipeline-ready 3D outputs Editable 3D worlds for Unity and Unreal Engine, with standard 3D exports including mesh, 3DGS, and point clouds. 🔹 Unified world model system One model family for world generation and reconstruction across synthetic and real-world scenes. 🔹 Interactive character mode Explore generated 3D worlds in real time with physics-aware movement and collision support. ✨ Apply for access: 3d.hunyuan.tencent.com/sceneTo3D 🔗 GitHub: github.com/Tencent-Hunyua… 🤗 Hugging Face: huggingface.co/tencent/HY-Wor… 📄 Technical Report: 3d-models.hunyuan.tencent.com/world/world2_0…

@cocktailpeanut Not to bash on anyone, but conclusion from your demo: - nothing special - not as advertised - can be achieved with outpainting on previous generation, save previous generations in sequence, map them in a 2d pane humm, anything with AI in the name these days, I guess.

OVIE Web UI I wanted to try out OVIE but turns out the official github repo is just a jupyter notebook, so I asked AI in Pinokio (Codex) to build a UI. 20 minutes later, a fully functional gradio app to play with OVIE. Enjoy. Works on ALL OS (Mac, Windows, Linux)

@LadyHUNK4dbd2 @TheRealQuenny @sean_gause

@Sneaky2x @TheRealQuenny @sean_gause you're gay

@nvidia If you cannot deliver good quality the price of your tokens is 0 Even if it costs a million dollars to run the model. So, Tokens are NOT a good metric at all. + A Token is a variable, it can be multiple characters and different characters all the time even on the same prompt.

@nvidia Actually, no it's not the metric that should be receiving most attention. The quality of your entire output vs the consumption of that compute is the metric that you should be paying attention. Tokens don't matter if you can never get good quality.

🧵Why This Is the One Business Metric That Matters Most

@NoahKingJr The longer I watch this, the more I think that if they bash their heads on the button they might have higher chance of winning.

Me liking every post that has "grok" just to see the new animation

@vikhyatk In a chain of events, this hard decision had to be made. AI Contest Ruby will truly be missed. 16 whole years of pure ruby entangled around two files, mesmerizing. RIP 💀

because of AI, it is no longer sustainable for us to operate as an open source business after much consideration, i have decided to close source the ruby starter pack for google ai contest 2010

@ForwardFuture Do you hear how many words were needed to actually attempt to extract relevant information? Yep, people will just disconnect from those noisy channels. The solution is more spam preventions and account verifications: subsequently flagged as AI, ban.

As AI content takes over the internet, what's the future of web browsing? Aravind Srinivas, CEO of Perplexity says "Whenever I post a tweet on X, a ton of responses are bots. I keep marking them as spam. And it seems pretty hard to fight." "There's a decoupling between the human and the actual internet. Now there's going to be an agent in the middle." "Go read this article for me. Filter all the ones that look like AI and spam. Pull up all the signal, summarize it, give it to me in the format I want to ingest."

@systemdesignone I get the take, but that's the same thing as saying don't use a Nail Gun because you will lose your Nail Hammering skills.

You need to code more, and you need to use less AI. You need to set aside time to train your problem-solving ability, and you need to give it time to struggle. Especially now that it's so easy for your skills to atrophy. Refuse to become intellectually dependent.