
Quilibrium Community
838 posts

Quilibrium Community
@QuilibriumOne
Quilibrium is building a better Internet: private, scalable and decentralized. This is the community account, not managed or endorsed by Quilibrium Inc.


Who cares about words anyway ...

I talked to two Israeli sources on why Iranian launches continue to increase, despite US-Israeli claims that they have destroyed almost all of the launchers. Here is what they said: 1) The 90–95% drop in volume claimed by CENTCOM earlier in the month was probably a temporary lull as Iran repositioned its remaining launchers into hardened sites. Independent satellite analysis suggests that a significant portion of the "80% destruction" claimed by the IDF actually hit high-fidelity decoys. 2) Despite fewer launchers, the lethality per strike has increased. Iran's shift to cluster warheads has allowed a single missile to impact multiple locations simultaneously, compensating for the lower volume of launches 3) Iran has successfully set up mobile, underground units able to fire at steady rates. Iran used that quiet period to move their remaining ~100-120 heavy launchers into "Super-Hardened" facilities 4) Iran is utilizing its Zolfaqar and Dezful road-mobile launchers. These units move from hardened tunnels to pre-surveyed launch spots, fire, and return underground in under 10 minutes, often before coalition drones can re-task for a strike. 5) Because these launching units are decentralized, it is very hard for US and Israeli intelligence to get info on them. Israel and the United States do not have an answer to this problem. That is why they are trying escalation on energy sources instead. But that is backfiring.




🇷🇺RUSSIA HACKERS TARGET SIGNAL & WHATSAPP … NO HACK NEEDED Dutch intelligence says Russian state hackers are running a global campaign to hijack Signal and WhatsApp accounts used by officials, military personnel, and civil servants. The trick? Pretend to be “Signal Support” and simply ask users for their verification codes. No software exploit required. Once inside, attackers can quietly read private chats and group messages. Encryption works. Human error… less so. Source: Dutch Government

@cass_on_mars shipped some pretty impressive updates (Klearu and MetaVM) for @QuilibriumInc. In case it was too complex for you, here it is explained for normies! MetaVM — What is it? Imagine you hire someone to do a maths problem for you. Normally, you'd have to redo the entire problem yourself to check they got it right. MetaVM is a tool that lets that person hand you a tiny "receipt" that mathematically proves they did the work correctly, without you having to redo anything. One quick check and you know the answer is legit. What makes it stand out is that it works across three major computing worlds: RISC-V (a general purpose chip architecture, the kind that runs Linux, and the direction Vitalik has publicly said he wants to move towards), Ethereum's EVM (the engine behind ETH smart contracts), and Solana's BPF (the engine behind Solana programs). It can verify an entire Ethereum block or an entire Solana slot in one go. It also plugs into Quilibrium's own cryptographic foundation, while remaining compatible with Ethereum's. That means it can speak both languages. Why it matters: For Quilibrium's network, this solves the fundamental trust problem. You don't have to trust the random machine that ran your code. You just check the proof. That's what makes decentralised computing actually work rather than just being a nice idea. The bigger picture: Quilibrium's founder has already gone directly to Vitalik pointing out that MetaVM does exactly what Ethereum's own roadmap calls for. And because the code is released under strict rules (AGPL), any company that wants to build a business on top of it would either have to make their entire product open and free, or come to Quilibrium for a commercial deal. That's a deliberate move to prevent big players from taking the technology, privatising it, and extracting value without giving anything back, which is exactly what happened with Ethereum's own codebase when companies like Coinbase built Base on top of it. It positions Quilibrium not just as a technology provider but as a gatekeeper against corporate value extraction in the crypto ecosystem. Klearu — What is it? Right now, when you use ChatGPT or any AI chatbot, you're sending your raw thoughts, questions, and data straight to a company's servers. They can see everything you type. Klearu is Quilibrium's answer to that problem. It lets you use AI models without anyone seeing what you actually asked. It has two big pieces: The first is speed. Normal AI models process everything: every single connection in the neural network fires for every word, even when most of that work is pointless. Klearu uses a technique based on peer reviewed research (the SLIDE paper family) that flips the approach. Instead of brute forcing through the entire model, it uses smart shortcuts to figure out which tiny fraction of the network actually matters for your specific input and only runs that part. The result, proven in academic benchmarks, is that a regular CPU can outperform expensive GPU hardware for certain workloads. That's a massive deal because Quilibrium's network runs on regular computers, not GPU farms. The second is privacy. Klearu lets two parties work together on AI inference where neither side sees the other's secrets. The person running the model never sees your prompt. You never get access to the model's weights. Both sides work on encrypted data the whole time, using real cryptographic protocols, not "trust us, we deleted the logs" promises. The maths guarantees it. Everyone is talking about AI right now, but almost nobody is solving the privacy side. Every big AI company has full access to every conversation you have with their models. Klearu means $QUIL could offer AI as a service where privacy is baked into the maths itself. A node operator on the network could run a LLaMA model and serve your requests without ever knowing what you asked or what the answer was. That doesn't exist anywhere else in a meaningful form right now. The trade off: This is still early. The privacy mode adds real overhead, roughly 2MB of encrypted back and forth per token at the highest security level, which is heavy. And the benchmarks so far use smaller models (up to 1.7 billion parameters), not the massive models people associate with frontier AI. But as a foundation for private AI on a decentralised network, it's one of the most technically serious attempts out there. And crucially, it runs on CPUs, which means it's built to work on the kind of machines that already power the Quilibrium network rather than requiring expensive specialised hardware that would centralise everything again. At a $15M valuation, $quil has one of the best R/R setups for me personally! Don't be lazy, guys, spend some time researching. Q! Tagging @AlgodTrading since he's the guy who always looks for conviction plays.

A lot of the "private" AI options out there are lying to you. Look under the hood. Use the dev tools on the browser to see what is actually sent. It's plaintext. What they're actually doing is _promising_ they won't look. We don't do that. See for yourself. We have an inspector tab so you can see the actual traffic data, but you can confirm it in the browser. klearu-demo.qstorage.quilibrium.com Don't trust. Verify.

MetaVM has been released. Prove execution of RISC-V, EVM, and Solana sBPF in ZK. Supports Quilibrium's BLS48-581 and Ethereum's BLS12-381 natively, without needing a GPU. MetaVM's RISC-V compatibility is the first full ZK RV64IMAC instruction set – you can run Linux in MetaVM, and prove everything that happened within the VM session, prove a block on Ethereum, or a slot on Solana, and emit proofs compatible with either Q or Ethereum. github.com/QuilibriumNetw…





okay so > aws is down > half the l2s went with it only one question we should have is: who’s building the decentralized aws?


MetaVM has been released. Prove execution of RISC-V, EVM, and Solana sBPF in ZK. Supports Quilibrium's BLS48-581 and Ethereum's BLS12-381 natively, without needing a GPU. MetaVM's RISC-V compatibility is the first full ZK RV64IMAC instruction set – you can run Linux in MetaVM, and prove everything that happened within the VM session, prove a block on Ethereum, or a slot on Solana, and emit proofs compatible with either Q or Ethereum. github.com/QuilibriumNetw…

Today, we are publishing one of the side tracks of research ongoing with Q, our E2EE ML training and inference library, klearu: github.com/QuilibriumNetw… SLIDE proved that hash tables can beat GPUs at training deep networks. Further works compounded on this, and Klearu is the first native Rust implementation built on top of this research, extending it to LLM inference, sparsity prediction, and private two-party computation. In the current days we're seeing deeper trust being placed on AI, while the largest of providers are collecting this data for the purpose of not only training, but also advertising, or even selling this data to others. The risks grow worse with every passing day. The majority of AI research for private AI exists in the form of using TEEs – but we've seen time and time again that using TEEs for privacy is disastrous, guaranteed to leak, and even by it's name, is a massive requirement of trust. Outside of this, other private AI looks towards FHE. We know, at least for the near future, that FHE cannot perform at a speed high enough to be generally useful. So instead, we adopted 2PC, with flexible security configurations, where users can be assured that their requests remain private. The majority of these research projects have strictly an output of papers, with no or limited real world instances of their use. Klearu's implementation is available now, with simple instructions for developers to try it out.

/klearu is out github.com/QuilibriumNetw…

Today, we are publishing one of the side tracks of research ongoing with Q, our E2EE ML training and inference library, klearu: github.com/QuilibriumNetw… SLIDE proved that hash tables can beat GPUs at training deep networks. Further works compounded on this, and Klearu is the first native Rust implementation built on top of this research, extending it to LLM inference, sparsity prediction, and private two-party computation. In the current days we're seeing deeper trust being placed on AI, while the largest of providers are collecting this data for the purpose of not only training, but also advertising, or even selling this data to others. The risks grow worse with every passing day. The majority of AI research for private AI exists in the form of using TEEs – but we've seen time and time again that using TEEs for privacy is disastrous, guaranteed to leak, and even by it's name, is a massive requirement of trust. Outside of this, other private AI looks towards FHE. We know, at least for the near future, that FHE cannot perform at a speed high enough to be generally useful. So instead, we adopted 2PC, with flexible security configurations, where users can be assured that their requests remain private. The majority of these research projects have strictly an output of papers, with no or limited real world instances of their use. Klearu's implementation is available now, with simple instructions for developers to try it out.



