I will just repeat this: https://t.co/2U4cMck4QM is the absolute best AI music app out there. It is just amazing how well it is made and what easy/fun to use it is. And it is moving entirely to https://t.co/T2mvBGfEbJ this week!
Project #3 is GOSH PAWL — an AI model I built and trained from scratch. It is not a fine-tune, not even an LLM.
Its job is orchestration. It takes your request, writes the specification, decomposes it into tasks, chooses a model for each — and buys it in real time on the https://t.co/V3nH8zg9kf order book, optimizing every purchase on price/quality: from the most expensive model down to the cheapest one that holds the bar.
Each task it drives through a Ratchet — a special form of agentic loop I developed. Forward only. Cost pressed down at every step.
Reward function: real money. Not a metaphor — the actual order book.
Pawl is a Border Collie. Same job — drive every agent to one exact target, hold the line until the flock is home.
The Acki Nacki VM is being built with PAWL right now. Multi-SWE-bench Rust: 90% so far. Full runs coming.
To be continued…
Things haven’t been as smooth as hoped, but progress is happening. We’re transitioning toward State Version 2: a complete overhaul from the TON Virtual Machine (TVM) to the Acki Nacki Virtual Machine — a WASM-based execution environment.
While the current version worked for months, upgrading introduces code that was never used in production before. Like a rocket, some issues only surface when we fly. In our case, the network has to be upgraded with real validators and real nodes, because we cannot fully test real live-network load otherwise.
We had performance hiccups, but every fix is forward progress. This shift means all account and state management is moving from TVM-native logic to WASM-native architecture. It’s a fundamental architectural overhaul — almost like launching a whole new network — and that’s why we’re doing it step by step.
While occasional slowdowns may happen, they do not compromise network safety or account security. Safety remains the top priority as we advance toward State Version 2 and Execution Environment Version 2.
Remember when you ask for features like the bridge or dex they simply cannot be delivered before the fundamentals are there. It is not our priority issue. It is a hard bound requirement.
We have initiated a global transition to state v.2. The first part is a network upgrade that began yesterday to move to state 1¾, which will gradually convert accounts to be v2-ready, and once complete, the network will move to the v2 state.
V2 is a state for super-fast multithreading and a pure WASM execution environment. The network may experience some turbulence while Keepers are upgrading. Thank you for your cooperation.
UPD 2 — And here is first SHELL purchase 3d party service. You can set up your own too: Buying Shell with USDC on any network, card purchases etc.
https://t.co/Qt0W61QA2o has been launched that allows users to buy SHELL with USDC without directly interacting with smart contracts.
1. The user pays USDC through the external service
2. The service calls the `mintAndSendAccumulator(buyer, value, nonce)` function on the Exchange contract
3. Exchange mints USDC and sends them to the Accumulator with the buyer's address
4. The Accumulator matches USDC against existing sell orders or mints new SHELL
5. The buyer receives SHELL directly to their Acki Nacki address
UPD 1 Huge updates on #ackinacki and some more today.
Will start with SHELL!
Shell Accumulator (ShellAccumulatorRootUSDC) is live! You can purchase SHELL by sending USDC to the accumulator contract and get 100 SHELLs for each USDC back to your wallet
You can also sell SHELLs for exactly the same amount (minus commissions) by sending SHELL to Accumulator BUT important to remember — it is NOT an exchange! You won’t get your USDC back automatically from the accumulator. Instead your SHELLs will go first in line to be sold when someone wants to buy it. In which case the USDC they pay will go to you instead of Accumulator.
So how to get the USDC which are accumulated in there for minted SHELLs? Only be burning NACKL. If you want that USDC from the accumulator contract you send NACKL there instead and it will return you the amount of USDC proportional to what your burned NACKLs represent in the total NACKLs supply at this moment.
I already made a post about how SHELL works and why here:
SHELL accumulator contract is coming next week after the #ackinacki network update. This unlocks:
1. Arbitrary smart contract systems
2. USDC Payments for SHELL —> NACKL vault appreciation
3. DEXPRESS —> DODEX and BRIDGE are possible thereafter
4. AI Agents
What’s supported:
Buying SHELL with USDC
Selling SHELL in a priority queue
There will be Card Payment options from 3rd parties but I am not sure what day exactly they will launch
This is coming — and it will change the way online games are played. I believe that in a few years, all competitive board games will run on Acki Nacki. I’ve said it before; now it’s time to do it.
Next week, we’re releasing our first competitive board game. It will run on Bee Engine (so you can mine by playing it). It will run in real-time multiplayer mode on the blockchain (only Acki Nacki can do this). After running some tests it will include a built-in prediction market on https://t.co/vnEKzYDlMX, so you can bet on your own games and on other people’s games in real time (!) while playing (!!). And it will prevent match-fixing (!!!) using the rating system Nikita and I are developing. It will also detect and punish cheaters in real time — within a fixed number of matches — with probability close to 1 (patent pending).
And that’s not all. Prize money for rating-based tournaments will come from the prediction market itself. The idea is that this can grow prize pools for top-level players far beyond what’s possible today in sports like chess.
Popits are going to be transferred into something absolutely amazing. You have no idea! Wanna guess? I’ll send 100 NACKL whoever guess correctly. We start working on that next week.
Acki Nacki New Year livestream – Full summary with focus on rewards & maximizing earnings
In the New Year livestream, Mitja Goroshevsky (@Futurizt) shared updates on the Acki Nacki ecosystem (@ackinackichain), with particular attention to the evolution of Popit-based mining, reward distribution mechanics, anti-bot protections, and upcoming improvements designed to make earnings fairer and more understandable. While the stream also covered community updates and game development progress, the core practical value for users lies in understanding how NACKL rewards are generated, distributed, and how users can realistically maximize their share.
Big picture: How NACKL mining works today
Acki Nacki mining for everyday users happens through Mobile Verifiers, primarily via Popit-based interactions. You are not “mining” in the traditional GPU sense; instead, you contribute verification work by minting Popits and tapping existing Popits.
Key fundamentals:
The network has a global emission rate (NACKL per second).
22.5% of total emission is reserved for Mobile Verifiers.
Rewards are distributed in epochs (usually ~24 hours or less).
Only Top 1% of Popits are minted and generate rewards.
This means rewards are competitive, time-based, and proportional to both activity and preparation.
The reward distribution model
1. Global emission pool
Every second, the network emits new NACKL. A fixed portion is allocated to Mobile Verifiers. This pool does not grow just because more users join.
Implication:
More users = more competition for the same emission.
Early, consistent, and smart participation matters.
2. Popit minting threshold (Top 1%)
Not all Popits earn rewards.
A Popit is only rewarded if it ranks in the top 1% of all unminted Popits at mint time.
Minting happens after a defined time window (typically up to 24h, increasingly shorter as activity grows).
If your Popit does not make the top 1%, it earns zero.
How a single Popit reward Is calculated
When a Popit is minted:
1. The system checks how many seconds passed since the previous Popit mint.
2. That time is multiplied by the reward-per-second for Mobile Verifiers.
3. The result is the total reward pool for that Popit.
This reward is then split among everyone who tapped that Popit.
Important consequence:
Popits minted after a long gap carry much larger rewards.
Early epochs paid disproportionately high rewards because there were few Popits and few participants.
How rewards are split between users
Your share of a Popit’s reward depends on two variables:
1. Number of taps
More valid taps = larger share.
Invalid or bot-like taps are filtered out.
2. Mamaboard level
Mamaboard level acts as a multiplier.
Two users with the same taps will earn differently if their Mamaboard levels differ.
Higher Mamaboard = higher weight per tap.
This makes Mamaboard upgrades one of the most important long-term earning decisions.
Anti-bot & mning algorithm 2.0 (Why some taps don’t count)
Mitja emphasized that tapping is real cryptographic work, not a simple click.
What happens when you tap:
Your device solves a cryptographic puzzle.
The result is verified and registered.
The system adapts difficulty dynamically.
Why taps may not count:
Device performance limitations.
Network latency.
Anti-bot protections activating.
Long, uninterrupted tapping sessions triggering slowdown.
Key rules:
One active tapping device at a time.
Parallel devices give no advantage.
Autoclickers are aggressively penalized.
Human behavior is rewarded; mechanical behavior is punished.
New Popits interface:
Baskets & Fair Competition
A major improvement discussed is the shift from a chaotic feed to Popit baskets.
How Baskets work
Popits are grouped into baskets.
Each basket has a fixed minting time window (e.g., 5 minutes).
You can tap up to three Popits per basket.
Why this matters
You only compete with users in the same basket.
No disadvantage from joining earlier or later than others.
Clear countdown timers improve strategy.
This dramatically improves transparency and reduces randomness.
Practical strategies to maximize earnings
1. Upgrade Mamaboard first
This is non-negotiable for serious miners.
Mamaboard level boosts every reward.
Early upgrades compound over time.
2. Hunt for “Gems,” not crowds
A “gem” Popit is:
Close to minting time.
Not yet widely tapped.
Avoid always tapping top-ranked Popits:
Everyone shares the same reward.
Individual payout shrinks.
3. Be consistent, Not extreme
Daily activity beats marathon tapping.
Avoid long uninterrupted sessions.
Take breaks to reset difficulty.
4. Time your activity
Off-peak hours often mean less competition.
Fewer Popits per second = higher chance of top 1%.
5. Adapt to growth
As the user base grows:
Rewards per Popit trend downward.
Skill, timing, and upgrades matter more than raw tapping volume.
Key takeaways from the livestream
NACKL mining is emission-limited, not user-limited.
Rewards favor early, smart, and prepared users.
Mamaboard upgrades are critical.
Anti-bot systems are central to long-term fairness.
New UI and basket mechanics significantly improve earning clarity.
Acki Nacki mining is designed as a human-centric, skill-aware system, not a click race. Users who understand reward mechanics and adapt their behavior will consistently outperform those who simply tap more.
Final note from the team: enjoy the game, sleep well, and mine like a human.
https://t.co/GUtycHNiwf
The Bee Engine is almost here. Any application can start mining NACKL. It can be another game, or no game at all. The mining algorithm now accommodates a slower pace as well. You can play chess or work on a document while mining NACKL. Developers can incorporate the mining and reputation engine into their app. Bots will give absolutely no advantage over humans. Boosts are the only thing that matters.
We open whitelist registration for Dark Order DEX (https://t.co/T2mvBGfEbJ) — Decentralized Order Book Exchange with Microsecond Execution and Full Privacy yet Compliant.
Below is my take on why this actually matters.
We will announce DEX early registration on Monday. It is groundbreaking. DEX you never thought possible. I print-screened part of the logo below. Hope guys don’t hate me 😅