Author: Kuri, Deep Tide TechFlow

Recently, the crypto market seems to be warming up, and some leading projects have started innovating, like the long-silent NFT.

On May 1st, an NFT series called Slonks was launched on Ethereum, with a mint price of less than 0.004 ETH, equivalent to less than seventy yuan.

Today, six days later, the floor price is 0.123 ETH, an increase of about 60 times. The 7-day trading volume on OpenSea is 586 ETH, with over twenty-three thousand transactions. The total supply is ten thousand pieces, and 1348 pieces have already been permanently burned, leaving 8642 pieces in circulation.

What level do these numbers represent in the NFT market of 2026? In the same week, the eight-year-old blue-chip CryptoPunks had only 20 transactions. Slonks had twenty-three thousand.

And tonight at nine o'clock, this project will also release a token called $SLOP.

If you search for Slonks on Twitter, you'll find an interesting phenomenon. Almost everyone is saying this project is "mechanically clever" and "design is self-consistent," but when you ask them to explain how it works, most people get stuck after saying "images can be exchanged for tokens, and tokens can be exchanged for images"...

Some say it is GameFi, an on-chain AI art experiment, or an evolved version of CryptoPunks; others say it is essentially a "dreamlike game of baby-hunting."

After a quick research, I think these statements touch on the subject, but none seem to hit the point.

The most interesting aspect of Slonks may be turning a very counterintuitive idea into a business. The idea is:

What AI draws wrong is worth more than what it draws correctly.

What makes AI's wrong pixels valuable?

This project differs from previous NFTs in that its NFTs are not image files.

Traditional NFTs typically store a completed image on an off-chain server, with only a link placed in the contract. Slonks is different; it directly embeds an AI image generation model into an Ethereum smart contract, totaling only 214KB, roughly the size of a low-resolution mobile wallpaper.

Every time someone views a Slonk, the contract executes the model inference on the spot, generating the image in real time.

No image is stored, only the capability to draw. This is actually a micro-innovative NFT gameplay.

So, what is this model actually drawing?

It's essentially copying... the model's task is to draw a copy of each of the ten thousand original images of CryptoPunks. Each CryptoPunk corresponds to a Slonk, with the model looking at the original image and trying to redraw it using the same color palette.

However, a 214KB model struggles to remember ten thousand faces.

Each image consists of 576 pixels, and on average, the model will misplace about 24 pixels, meaning each image has about 4% of its elements differing from the original. Out of ten thousand, only 32 images have been perfectly replicated, and the rest all have varying degrees of "distortion."

The project refers to these misdrawn, distorted pixels as slop.

If no pixels are wrong, slop is 0; if all are wrong, slop is 576. The project's developer Hirsch summarized this project's attitude on Twitter with one sentence:

The slop is not a bug. It is the medium.

This means that the AI model has re-copied the old NFT OG project, but due to its errors, the different wrong appearances create different scarcity and speculative values.

Thus, the entire economic model of the project is built on this logic: the more wrong it is, the more valuable it is.

Slonk holders can perform an action called Merge. Take two Slonks of the same level you have, keep one, and burn the other. The contract will mix the features of the two images together and let the model redraw a new one.

Because two different images are mixed together, the result will differ even more from the original CryptoPunk, not less.

With each Merge, slop can only increase. Moreover, the burned one can never be recovered, reducing the total supply by one.

This is why, just six days after its launch, over one thousand three hundred pieces have already been burned from the ten thousand. Players keep merging to create versions with higher slop while reducing the circulating amount.

By this point, Slonks has become quite an interesting on-chain art experiment, effectively creating an artificial scarcity using AI algorithms.

$SLOP: Pricing Each Wrong Pixel

What transforms this project into a real business is likely the upcoming official token $SLOP.

It aims to turn the number indicating "how many pixels were drawn wrong" on each image into something that can be bought and sold.

How does it do this? Through an operation called Void.

Holders can "send their Slonk into the Void," removing the NFT from circulation; the contract will mint an equal amount of $SLOP tokens based on this image's slop value, on a 1-to-1 basis.

For example, if your image's slop is 287, you will receive 287 $SLOP. If slop is 450, then that's 450 tokens.

This process cannot just be arbitrary; the contract requires you to generate a ZK proof first, proving that the on-chain model's rendering result of this image is indeed as you claim, and only then will it mint the tokens.

In simple terms, you have to first "validate" that your image is indeed wrong by that many pixels for the contract to acknowledge it.

The Slonk sent into the Void has not been destroyed; it still lies in the contract. This leads to a second operation: Revival.

By spending a certain amount of $SLOP, you can randomly draw a Slonk back from the Void. You cannot choose which one you get, and once this image returns, it will be redrawn by the model, likely looking different than before.

The pricing for revival uses a Dutch auction: starting at 576 $SLOP, decreasing by 1 for each block, down to a minimum of 100. But as soon as anyone buys once, the price immediately bounces back to 576 and starts decreasing again. If you want to draw at a cheaper price, you'll have to wait, but during that time, others might jump in first.

Most revived images will closely resemble the original CryptoPunk with not much slop. However, there is a 1% probability that the model will "completely go wild" and generate an extreme version with slop over 400. This 1% chance is like a lottery.

By now, you might have noticed that Void and Revival form an economic cycle of "图币互转" (image-token conversion):

1. Buy a high slop Slonk at a low price and send it into the Void for tokens.
2. Wait for the tokens to rise, then spend those tokens to revive a new image.
3. If lucky, you draw a high slop one, then send it back into the Void for more tokens.
4. Images turn into tokens, tokens turn into images, back and forth.

The project also designed an automatic flywheel: the $SLOP trading pool collects a 2% fee, half of which will be automatically used to buy Slonks at the floor price from OpenSea and then send them into the Void as future revival inventory.

The more active the trading, the more inventory in the Void, and thus the more available for revival draws.

Regarding tonight's token launch, Hirsch announced the specific process a few hours ago. 576,000 SLOP will be placed in a unidirectional liquidity pool, with an initial market value of about 50 thousand dollars.

Buyers need to bring their own ETH to purchase. For the first six hours post-launch, only token buying and selling will be allowed; no Void operations can be executed. This means these six hours are purely a price discovery period, allowing the market to set a price for $SLOP before opening up the Void, fully establishing the conversion channel between NFTs and tokens.

The hard cap for $SLOP is 5,760,000, which exactly equals ten thousand images multiplied by 576 pixels each. However, this number will never be fully minted since images burned in Merge do not produce any tokens, and those more than a thousand already burned have their slop eternally locked.

I think the smartest aspect of this design is that it binds "speculation" and "creation" to the same action.

When you Merge two images, you are speculating because a higher slop means more potential tokens; but you are simultaneously creating because the model will draw a brand-new, never-before-existed 24x24 pixel image. You earn money and engage in on-chain art using the same button, and they mutually reinforce each other.

Thus, this gameplay can be seen as a micro-innovation Ponzi economic structure for NFTs, creating scarcity, image-token exchange, and designing deflation.

An Old Engraving Veteran, Revamping the Old

As for the creator of Slonks, Michael Hirsch, not many in the community know him, but he is not a novice.

Do you remember ETHS on Ethereum during the previous excitement of engravings? The project had a peak market value of about 420 million dollars, and Hirsch is the founder of that project.

After the engraving trend cooled down, he turned to create Blockhash, a small studio focusing on on-chain products, developing DEXs, NFT trading platforms, token-gated chat tools, and various things he describes as "strange crypto experiments."

Slonks is the latest work from this studio.

A person who has experienced the ups and downs of the engraving cycle chooses to launch a project with a very high understanding threshold now; this itself is a noteworthy signal. Has liquidity and sentiment on-chain warmed up, prompting some micro-innovations to start speculation?

I think we still need to observe.

As mentioned before, tonight at nine o'clock, the $SLOP token will launch. For the first six hours after launch, only trading of the token will be allowed; no Void operations can exchange NFTs for tokens. These six hours are purely a price battle period, with no new supply and buyers setting prices among themselves.

After six hours, the Void channel will open and the two-way conversion between NFTs and tokens will officially start.

The initial market value of fifty thousand dollars implies extremely thin liquidity, and early price fluctuations will be very severe. The white paper still shows "v1 · draft" at this stage.

This project has technical ingenuity, a proven founder, and a self-consistent economic cycle logic, but it essentially trivializes micro-innovation on the economic model and artificially creates scarcity.

People who have experienced several cycles will likely smile knowingly at this gameplay, but in any case, having new things in this current market is a good trend in itself.

DYOR.