Taho Early Roller 🎲 rolly.io retweeted
CLAIM. LIVE.
claimyshare.io/tge
72 hours until $CYS is sent to your wallet.
Taho Early Roller 🎲 rolly.io retweeted
Maximize your share of the $30K Early Rollers Race pool with a strong wallet.
Your wallet isn't just a login.
It directly affects your leaderboard spot and your final rewards.
Here's how Wallet Scoring works 👇
Taho Early Roller 🎲 rolly.io retweeted
Anh chàng này trong video 16 phút này dạy bạn cách kiếm tiền từ một kênh YouTube chỉ trong 9 ngày
Tôi đã vietsub lại cho anh em không biết tiếng anh dễ hiểu
Thêm vào dấu trang đi, nó sẽ rất hữu ích cho bạn.
Taho Early Roller 🎲 rolly.io retweeted
Agentic WaaP is live.
Compromise the agent. Steal nothing.
You don't have to give your private key to an agent, nor to your wallet provider.
Taho Early Roller 🎲 rolly.io retweeted
a new paper by our co-founder @fede_intern and @diego_aligned. they introduced a new way to achieve practical verifiable ai.
we are brainstorming if this could be added to improve or build new products to our stack. we also need a new whitepaper tshirt for diego :)
LLMs now make critical decisions in hospitals, defense, banks, and governments. Yet nobody can verify which model actually ran, or whether the output was tampered with. A provider or middleman can swap weights, silently requantize the model, alter decoding, inject hidden prompts, do supply chain attacks, or change the deployment surface without the user knowing.
This problem is already serious. It will become critical.
We think this needs a practical solution, not just a theoretically clean one. CommitLLM is designed to be deployable on existing serving stacks now: the provider keeps the normal GPU serving path, does not need a proving circuit, does not need a kernel rewrite, and does not generate a heavy proof for every response.
In practice, two families of approaches dominated the conversation before this work: fingerprinting, which can be gamed, and proof-based systems, which are theoretically strong but too expensive for production inference.
We built CommitLLM to target the middle ground.
The core idea is to keep the verification discipline of proof systems, but specialize it to open weight LLM inference. The cryptographic core is simple: Freivalds style randomized checks for the large linear layers, plus Merkle commitments for the traced execution. Then a lot of engineering work is needed to make that line up with real GPU inference.
The key trick is this. A provider claims `z = W × x` for a massive weight matrix. Normally you would verify that by redoing the multiply. Instead, the verifier samples a secret random vector `r`, precomputes `v = rᵀ × W`, and later checks whether `v · x = rᵀ · z`. Two dot products instead of a full matrix multiply. In the current implementation, a wrong result passes with probability at most `1 / (2^32 - 5)` per check.
A full matrix multiply, audited with two dot products.
Most of the transformer can then be checked exactly or canonically from committed openings. Nonlinear operations such as activations and layer norms are canonically re executed by the CPU verifier. The one honest caveat is attention: native FP16/BF16 attention is not bit reproducible across hardware. CommitLLM verifies the shell around attention exactly, then independently replays attention and checks that the committed post attention output stays within a measured INT8 corridor. So attention is bounded and audited, not proved exactly.
That means the protocol already gives very strong exact guarantees on the parts that matter operationally most. If an audited response used the wrong model, the wrong quantization/configuration, or a tampered input/deployment surface, the audit catches that exactly. That includes things like model swaps, silent requantization, and provider side prompt or system prompt injection.
Today the implementation and measurements are strongest on Qwen and Llama. But the protocol itself is not meant to be Qwen or Llama specific: we expect it to generalize across open weight decoder only families. What still has to be done is the engineering work to integrate and validate more families explicitly, and we are already working on that.
On the measured path, online generation overhead is about 12 to 14% with the provider staying on the normal GPU serving path. The heavier receipt finalization cost is separate and can be deferred off the user facing path. The main systems costs are RAM and bandwidth, not proof generation.
The full response is always committed, but only a random fraction of responses are opened for audit. Individual audits are much larger, roughly 4 MB to 100 MB depending on audit depth. The important number is the amortized one: under a reasonable audit policy, the added bandwidth averages to roughly 300 KB per response.
After too many weeks without sleep, I’m proud to show what I built with @diego_aligned: CommitLLM. Thanks Diego for your patience. I've been calling you at random hours.
The code and paper still need some cleaning and formalization. We’re already in talks with multiple providers and teams that have cryptography related ideas on how to improve it even more. We’re really excited about this and we will continue doubling down on building products in AI, cryptography and security with my company @class_lambda.
If governments, hospitals, defense and financial systems are going to run on LLMs, verifiable inference is not optional. It is infrastructure.
I will be explaining this in more details in the days to come and I will show how to test it and run it.
Taho Early Roller 🎲 rolly.io retweeted
Ethereum is ready to eat finance. Aligned makes it inevitable.
Aligned is the solution that can bring the world's businesses onchain. Choosing Aligned is choosing Ethereum as the backbone of global finance.
Join us 👇
Taho Early Roller 🎲 rolly.io retweeted
It’s officially live 🚀
The DAVE Labs Community Hub is where action turns into rewards.
Do simple tasks, earn points, and climb your way up the leaderboard.
The more you show up, the more you unlock.
Early users always have the edge.
Jump in now: hub.davelabs.io 🔥
Taho Early Roller 🎲 rolly.io retweeted
Introducing Grove Data 🌳🦜
Real time access to the metrics that matter.
A public dashboard with live, verifiable detail on Grove’s deployments onchain. Updated in real time as Grove evolves.
Trust through transparency.
Explore → data.grove.finance
Taho Early Roller 🎲 rolly.io retweeted
This level of traffic changes everything.
•350,000+ wallets scanned.
•500,000+ total users.
Real users.
Real activity.
Real demand.
Yes — we detected massive bot farms.
They are banned.
Leaderboard has been recalculated.
Now it reflects real numbers.
Tokenomics of CYS is protected.
We’ve implemented new registration protection.
Normal users won’t even notice.
Sybil farms definitely will.
When fake accounts attack you,
it means only one thing —
The project matters.
The team is working around the clock to secure the platform and keep everything stable.
We are actively monitoring all issues.
If you’re experiencing something —
Trust us.
We already know.
And it’s being handled.
This kind of organic + fake pressure combined
proves one thing:
CLAIM YOUR SHARE is needed.
The race continues.
Taho Early Roller 🎲 rolly.io retweeted
155,000+ wallets scanned.
220,000+ users joined in 10 hours.
And it’s still accelerating.
Season 1 just started.
The leaderboard is already moving every second.
2500 SOL.
12 seasons.
Positions are being taken right now.
Yes, traffic is insane.
Yes, we’re fixing bugs live.
That’s what real demand looks like.
While you’re reading this,
someone just passed you on the leaderboard.
Early users are stacking CYS.
If you wait — you compete from behind.
→ claimyshare.io/
Scan your wallet.
Claim your position.
Or watch others take it.
Taho Early Roller 🎲 rolly.io retweeted
CLAIM YOUR SOLANA is officially LIVE.
claimyshare.io/
Testing phase is over.
Official release has begun.
Season 1 is now active.
The leaderboard is open.
Rewards are real.
Emission has started.
You can now:
• Scan your Solana wallet.
• Earn CYS/SOL based on real on-chain activity.
• Open mystery boxes
• Compete for 2500 SOL.
• 12 seasons.
Early activity matters.
The earlier you join — the stronger your position.
No more beta.
No more testing.
This is the official launch.
Go now → claimyshare.io/
Scan. Earn. Compete.
Taho Early Roller 🎲 rolly.io retweeted
Total surveillance or total sovereignty?
The lines are being drawn right now.
Join our co-founder @king_kassam at Zenon Dubai for a new edition of @TradersBrawlDXB’s Money Monday!
He’ll be alongside @fractlCMOs discussing how we build a world where you actually own your data.
📅 Jan 26 | 7 PM - 11 PM
📍 Zenon Dubai
Taho Early Roller 🎲 rolly.io retweeted
Take control in your ANIMA chats.
AI shouldn’t be “one size fits all”.
Toggle between Chat, Research, and Plan modes to fit your flow.
Over time, ANIMA discovers who you are - and meets you where you’re at.
Powered by memory, context, and learning that only improves with time.
Taho Early Roller 🎲 rolly.io retweeted
GM 🌤️
What’s your go-to AI right now?
Everyone’s got their pick for their own reasons ✨
Daily-driving something else? Let us know!
59%ChatGPT
7%Claude
17%Gemini
17%Grok
1,206 votes • Final results Taho Early Roller 🎲 rolly.io retweeted
It’s time to stop switching tabs.
Swapping back and forth between ChatGPT and your project is so last year.
In ARC, it’s all unified.
✓ Your workflows
✓ Your data
✓ Your intelligence
You shouldn’t have to fight your tools just to get work done.
Taho Early Roller 🎲 rolly.io retweeted
ARC HAS BEEN BOOSTED ⚡️⚡️⚡️- 50x SPEED ENHANCEMENT
Multi-phased parallel instances deployed, with dynamic routing to ensure sub <100ms load across multiple services and api requests.
Taho Early Roller 🎲 rolly.io retweeted
Another year almost down, another step into the future.
Every year brings its own waves of adoption, and if BCG’s projection of 16T in assets moving onchain by 2030 holds, we’re closer than we think.
The coming shift feels inevitable. Let’s build for it together.


























