I've never seen this before.

Focus on native crypto development, not price activity or turmoil.

CT is boring now. Either folks are still angry about FTX or they're distracted by AI. Plus, it's year-end, and people rest for the holidays. 2022 was rough.

So DeFi fans can get inspired by something fresh. Who's building? As I read the Defillama daily roundup, many updates are still on FTX and its contagion.

I've used the same method on their Raises page. Not much happened :(. Maybe my high standards are to fault, but the business may be resting. OK.

The handful I locate might last us till the end of the year. (If another big blowup occurs.)

Hashflow

An on-chain monitor account I follow reported a huge transfer of $HFT from Binance to Jump Tradings.

I was intrigued. Stacking? So I checked and discovered out the project was launched through Binance Launchpad, which has introduced many 100x tokens (although momentarily) in the past, such as GALA and STEPN.

Hashflow appears to be pumpable. Binance launchpad, VC backers, CEX listing immediately. What's the protocol?

Hasflow is intriguing and timely, I discovered. After the FTX collapse, people looked more at DEXs.

Hashflow is a decentralized exchange that connects traders with professional market makers, according to its Binance launchpad description. Post-FTX, market makers lost their MM-ing chance with the collapse of the world's third-largest exchange. Jump and Wintermute back them?

Why is that the case? Hashflow doesn't use bonding curves like standard AMM. On AMMs, you pay more for the following trade because the prior trade reduces liquidity (supply and demand). With market maker quotations, you get a CEX-like experience (fewer coins in the pool, higher price). Stable prices, no MEV frontrunning.

Hashflow is innovative because...

DEXs gained from the FTX crash, but let's be honest: DEXs aren't as good as CEXs. Hashflow will change this.

Hashflow offers MEV protection, which major dealers seek in DEXs. You can trade large amounts without front running and sandwich assaults.

Hasflow offers a user-friendly swapping platform besides MEV. Any chain can be traded smoothly. This is a benefit because DEXs lag CEXs in UX.

Status, timeline:

Wintermute wrote in August that prominent market makers will work on Hashflow. Binance launched a month-long farming session in December. Jump probably participated in this initial sell, therefore we witnessed a significant transfer after the introduction.

Binance began trading HFT token on November 11 (the day FTX imploded). coincidence?)

Tokens are used for community rewards. Perhaps they'd copy dYdX. (Airdrop?). Read their documents about their future plans. Tokenomics doesn't impress me. Governance, rewards, and NFT.

Their stat page details their activity. First came Ethereum, then Arbitrum. For a new protocol in a bear market, they handled a lot of unique users daily.

It’s interesting to see their future. Will they be thriving? Not only against DEXs, but also among the CEXs too.

STFX

I forget how I found STFX. Possibly a Twitter thread concerning Arbitrum applications. STFX was the only new protocol I found interesting.

STFX is a new concept and trader problem-solver. I've never seen this protocol.

STFX allows you copy trades. You give someone your money to trade for you.

It's a marketplace. Traders are everywhere. You put your entry, exit, liquidation point, and trading theory. Twitter has a verification system for socials. Leaderboards display your trading skill.

This service could be popular. Staying disciplined is the hardest part of trading. Sometimes you take-profit too early or too late, or sell at a loss when an asset dumps, then it soon recovers (often happens in crypto.) It's hard to stick to entry-exit and liquidation plans.

What if you could hire someone to run your trade for a little commission? Set-and-forget.

Trading money isn't easy. Trust how? How do you know they won't steal your money?

Smart contracts.

STFX's trader is a vault maker/manager. One trade=one vault. User sets long/short, entrance, exit, and liquidation point. Anyone who agrees can exchange instantly. The smart contract will keep the fund during the trade and limit the manager's actions.

Here's STFX's transaction flow.

Managers and the treasury receive fees. It's a sustainable business strategy that benefits everyone.

I'm impressed by $STFX's planned use. Brilliant priority access. A crypto dealer opens a vault here. Many would join. STFX tokens offer VIP access over those without tokens.

STFX provides short-term trading, which is mind-blowing to me. I agree with their platform's purpose. Crypto market pricing actions foster short-termism. When you trade, the turnover could be larger than long-term holding or trading. 2017 BTC buyers waited 5 years to complete their holdings.

STFX teams simply adapted. Volatility aids trading.

All things about STFX scream Degen. The protocol fully embraces the degen nature of some, if not most, crypto natives.

An enjoyable dApp. Leaderboards are fun for reputation-building. FLEXING COMPETITIONS. You can join for as low as $10. STFX uses Arbitrum, therefore gas costs are low. Alpha procedure completes the degen feeling.

Despite looking like they don't take themselves seriously, I sense a strong business plan below. There is a real demand for the solution STFX offers.

Deep learning-based generative art is being researched. As usual, self-learning is better. Some models, like OpenAI's DALL-E 2, require registration and can only be used online, but others can be used locally, which is usually more enjoyable for curious users. I'll demonstrate the Stable Diffusion model's operation on a standard PC.

Let’s get started.

What It Does

Stable Diffusion uses numerous components:

• A generative model trained to produce images is called a diffusion model. The model is incrementally improving the starting data, which is only random noise. The model has an image, and while it is being trained, the reversed process is being used to add noise to the image. Being able to reverse this procedure and create images from noise is where the true magic is (more details and samples can be found in the paper).

• An internal compressed representation of a latent diffusion model, which may be altered to produce the desired images, is used (more details can be found in the paper). The capacity to fine-tune the generation process is essential because producing pictures at random is not very attractive (as we can see, for instance, in Generative Adversarial Networks).

• A neural network model called CLIP (Contrastive Language-Image Pre-training) is used to translate natural language prompts into vector representations. This model, which was trained on 400,000,000 image-text pairs, enables the transformation of a text prompt into a latent space for the diffusion model in the scenario of stable diffusion (more details in that paper).

This figure shows all data flow:

The weights file size for Stable Diffusion model v1 is 4 GB and v2 is 5 GB, making the model quite huge. The v1 model was trained on 256x256 and 512x512 LAION-5B pictures on a 4,000 GPU cluster using over 150.000 NVIDIA A100 GPU hours. The open-source pre-trained model is helpful for us. And we will.

Install

Before utilizing the Python sources for Stable Diffusion v1 on GitHub, we must install Miniconda (assuming Git and Python are already installed):

wget https://repo.anaconda.com/miniconda/Miniconda3-py39_4.12.0-Linux-x86_64.sh
chmod +x Miniconda3-py39_4.12.0-Linux-x86_64.sh
./Miniconda3-py39_4.12.0-Linux-x86_64.sh
conda update -n base -c defaults conda

Install the source and prepare the environment:

git clone https://github.com/CompVis/stable-diffusion
cd stable-diffusion
conda env create -f environment.yaml
conda activate ldm
pip3 install transformers --upgrade

Download the pre-trained model weights next. HiggingFace has the newest checkpoint sd-v14.ckpt (a download is free but registration is required). Put the file in the project folder and have fun:

python3 scripts/txt2img.py --prompt "hello world" --plms --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

Almost. The installation is complete for happy users of current GPUs with 12 GB or more VRAM. RuntimeError: CUDA out of memory will occur otherwise. Two solutions exist.

Running the optimized version

Try optimizing first. After cloning the repository and enabling the environment (as previously), we can run the command:

python3 optimizedSD/optimized_txt2img.py --prompt "hello world" --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

Stable Diffusion worked on my visual card with 8 GB RAM (alas, I did not behave well enough to get NVIDIA A100 for Christmas, so 8 GB GPU is the maximum I have;).

Running Stable Diffusion without GPU

If the GPU does not have enough RAM or is not CUDA-compatible, running the code on a CPU will be 20x slower but better than nothing. This unauthorized CPU-only branch from GitHub is easiest to obtain. We may easily edit the source code to use the latest version. It's strange that a pull request for that was made six months ago and still hasn't been approved, as the changes are simple. Readers can finish in 5 minutes:

• Replace if attr.device!= torch.device(cuda) with if attr.device!= torch.device(cuda) and torch.cuda.is available at line 20 of ldm/models/diffusion/ddim.py ().

• Replace if attr.device!= torch.device(cuda) with if attr.device!= torch.device(cuda) and torch.cuda.is available in line 20 of ldm/models/diffusion/plms.py ().

• Replace device=cuda in lines 38, 55, 83, and 142 of ldm/modules/encoders/modules.py with device=cuda if torch.cuda.is available(), otherwise cpu.

• Replace model.cuda() in scripts/txt2img.py line 28 and scripts/img2img.py line 43 with if torch.cuda.is available(): model.cuda ().

Run the script again.

Testing

Test the model. Text-to-image is the first choice. Test the command line example again:

python3 scripts/txt2img.py --prompt "hello world" --plms --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

The slow generation takes 10 seconds on a GPU and 10 minutes on a CPU. Final image:

Hello world is dull and abstract. Try a brush-wielding hamster. Why? Because we can, and it's not as insane as Napoleon's cat. Another image:

Generating an image from a text prompt and another image is interesting. I made this picture in two minutes using the image editor (sorry, drawing wasn't my strong suit):

I can create an image from this drawing:

python3 scripts/img2img.py --prompt "A bird is sitting on a tree branch" --ckpt sd-v1-4.ckpt --init-img bird.png --strength 0.8

It was far better than my initial drawing:

I hope readers understand and experiment.

Stable Diffusion UI

Developers love the command line, but regular users may struggle. Stable Diffusion UI projects simplify image generation and installation. Simple usage:

• Unpack the ZIP after downloading it from https://github.com/cmdr2/stable-diffusion-ui/releases. Linux and Windows are compatible with Stable Diffusion UI (sorry for Mac users, but those machines are not well-suitable for heavy machine learning tasks anyway;).

• Start the script.

Done. The web browser UI makes configuring various Stable Diffusion features (upscaling, filtering, etc.) easy:

V2.1 of Stable Diffusion

I noticed the notification about releasing version 2.1 while writing this essay, and it was intriguing to test it. First, compare version 2 to version 1:

• alternative text encoding. The Contrastive LanguageImage Pre-training (CLIP) deep learning model, which was trained on a significant number of text-image pairs, is used in Stable Diffusion 1. The open-source CLIP implementation used in Stable Diffusion 2 is called OpenCLIP. It is difficult to determine whether there have been any technical advancements or if legal concerns were the main focus. However, because the training datasets for the two text encoders were different, the output results from V1 and V2 will differ for the identical text prompts.

• a new depth model that may be used to the output of image-to-image generation.

• a revolutionary upscaling technique that can quadruple the resolution of an image.

• Generally higher resolution Stable Diffusion 2 has the ability to produce both 512x512 and 768x768 pictures.

The Hugging Face website offers a free online demo of Stable Diffusion 2.1 for code testing. The process is the same as for version 1.4. Download a fresh version and activate the environment:

conda deactivate  
conda env remove -n ldm  # Use this if version 1 was previously installed
git clone https://github.com/Stability-AI/stablediffusion
cd stablediffusion
conda env create -f environment.yaml
conda activate ldm

Hugging Face offers a new weights ckpt file.

The Out of memory error prevented me from running this version on my 8 GB GPU. Version 2.1 fails on CPUs with the slow conv2d cpu not implemented for Half error (according to this GitHub issue, the CPU support for this algorithm and data type will not be added). The model can be modified from half to full precision (float16 instead of float32), however it doesn't make sense since v1 runs up to 10 minutes on the CPU and v2.1 should be much slower. The online demo results are visible. The same hamster painting with a brush prompt yielded this result:

It looks different from v1, but it functions and has a higher resolution.

The superresolution.py script can run the 4x Stable Diffusion upscaler locally (the x4-upscaler-ema.ckpt weights file should be in the same folder):

python3 scripts/gradio/superresolution.py configs/stable-diffusion/x4-upscaling.yaml x4-upscaler-ema.ckpt

This code allows the web browser UI to select the image to upscale:

The copy-paste strategy may explain why the upscaler needs a text prompt (and the Hugging Face code snippet does not have any text input as well). I got a GPU out of memory error again, although CUDA can be disabled like v1. However, processing an image for more than two hours is unlikely:

Stable Diffusion Limitations

When we use the model, it's fun to see what it can and can't do. Generative models produce abstract visuals but not photorealistic ones. This fundamentally limits The generative neural network was trained on text and image pairs, but humans have a lot of background knowledge about the world. The neural network model knows nothing. If someone asks me to draw a Chinese text, I can draw something that looks like Chinese but is actually gibberish because I never learnt it. Generative AI does too! Humans can learn new languages, but the Stable Diffusion AI model includes only language and image decoder brain components. For instance, the Stable Diffusion model will pull NO WAR banner-bearers like this:

V1:

V2.1:

The shot shows text, although the model never learned to read or write. The model's string tokenizer automatically converts letters to lowercase before generating the image, so typing NO WAR banner or no war banner is the same.

I can also ask the model to draw a gorgeous woman:

V1:

V2.1:

The first image is gorgeous but physically incorrect. A second one is better, although it has an Uncanny valley feel. BTW, v2 has a lifehack to add a negative prompt and define what we don't want on the image. Readers might try adding horrible anatomy to the gorgeous woman request.

If we ask for a cartoon attractive woman, the results are nice, but accuracy doesn't matter:

V1:

V2.1:

Another example: I ordered a model to sketch a mouse, which looks beautiful but has too many legs, ears, and fingers:

V1:

V2.1: improved but not perfect.

V1 produces a fun cartoon flying mouse if I want something more abstract:

I tried multiple times with V2.1 but only received this:

The image is OK, but the first version is closer to the request.

Stable Diffusion struggles to draw letters, fingers, etc. However, abstract images yield interesting outcomes. A rural landscape with a modern metropolis in the background turned out well:

V1:

V2.1:

Generative models help make paintings too (at least, abstract ones). I searched Google Image Search for modern art painting to see works by real artists, and this was the first image:

I typed "abstract oil painting of people dancing" and got this:

V1:

V2.1:

It's a different style, but I don't think the AI-generated graphics are worse than the human-drawn ones.

The AI model cannot think like humans. It thinks nothing. A stable diffusion model is a billion-parameter matrix trained on millions of text-image pairs. I input "robot is creating a picture with a pen" to create an image for this post. Humans understand requests immediately. I tried Stable Diffusion multiple times and got this:

This great artwork has a pen, robot, and sketch, however it was not asked. Maybe it was because the tokenizer deleted is and a words from a statement, but I tried other requests such robot painting picture with pen without success. It's harder to prompt a model than a person.

I hope Stable Diffusion's general effects are evident. Despite its limitations, it can produce beautiful photographs in some settings. Readers who want to use Stable Diffusion results should be warned. Source code examination demonstrates that Stable Diffusion images feature a concealed watermark (text StableDiffusionV1 and SDV2) encoded using the invisible-watermark Python package. It's not a secret, because the official Stable Diffusion repository's test watermark.py file contains a decoding snippet. The put watermark line in the txt2img.py source code can be removed if desired. I didn't discover this watermark on photographs made by the online Hugging Face demo. Maybe I did something incorrectly (but maybe they are just not using the txt2img script on their backend at all).

Conclusion

The Stable Diffusion model was fascinating. As I mentioned before, trying something yourself is always better than taking someone else's word, so I encourage readers to do the same (including this article as well;).

Is Generative AI a game-changer? My humble experience tells me:

• I think that place has a lot of potential. For designers and artists, generative AI can be a truly useful and innovative tool. Unfortunately, it can also pose a threat to some of them since if users can enter a text field to obtain a picture or a website logo in a matter of clicks, why would they pay more to a different party? Is it possible right now? unquestionably not yet. Images still have a very poor quality and are erroneous in minute details. And after viewing the image of the stunning woman above, models and fashion photographers may also unwind because it is highly unlikely that AI will replace them in the upcoming years.

• Today, generative AI is still in its infancy. Even 768x768 images are considered to be of a high resolution when using neural networks, which are computationally highly expensive. There isn't an AI model that can generate high-resolution photographs natively without upscaling or other methods, at least not as of the time this article was written, but it will happen eventually.

• It is still a challenge to accurately represent knowledge in neural networks (information like how many legs a cat has or the year Napoleon was born). Consequently, AI models struggle to create photorealistic photos, at least where little details are important (on the other side, when I searched Google for modern art paintings, the results are often even worse;).

• When compared to the carefully chosen images from official web pages or YouTube reviews, the average output quality of a Stable Diffusion generation process is actually less attractive because to its high degree of randomness. When using the same technique on their own, consumers will theoretically only view those images as 1% of the results.

Anyway, it's exciting to witness this area's advancement, especially because the project is open source. Google's Imagen and DALL-E 2 can also produce remarkable findings. It will be interesting to see how they progress.

I apologize if this sounds basic, but throughout my career, I've always been clear that a company's activities are shaped by its founder(s)' morality.

I consider Palantir, owned by PayPal founder Peter Thiel, evil. He got $5 billion tax-free by hacking a statute to help middle-class savings. That may appear clever, but I think it demonstrates a shocking lack of solidarity with society. As a result of this and other things he has said and done, I early on dismissed Peter Thiel as someone who could contribute anything positive to society, and events soon proved me right: we are talking about someone who clearly considers himself above everyone else and who does not hesitate to set up a company, Palantir, to exploit the data of the little people and sell it to the highest bidder, whoever that is and whatever the consequences.

The German courts have confirmed my warnings concerning Palantir. The problem is that politicians love its surveillance tools because they think knowing more about their constituents gives them power. These are ideal for dictatorships who want to snoop on their populace. Hence, Silicon Valley's triumphalist dialectic has seduced many governments at many levels and collected massive volumes of data to hold forever.

Dangerous company. There are many more. My analysis of the moral principles that disclose company management changed my opinion of Facebook, now Meta, and anyone with a modicum of interest might deduce when that happened, a discovery that leaves you dumbfounded. TikTok was easy because its lack of morality was revealed early when I saw the videos it encouraged minors to post and the repercussions of sharing them through its content recommendation algorithm. When you see something like this, nothing can convince you that the firm can change its morals and become good. Nothing. You know the company is awful and will fail. Speak it, announce it, and change it. It's like a fingerprint—unchangeable.

Some of you who read me frequently make its Facebook today jokes when I write about these firms, and that's fine: they're my moral standards, those of an elderly professor with thirty-five years of experience studying corporations and discussing their cases in class, but you don't have to share them. Since I'm writing this and don't have to submit to any editorial review, that's what it is: when you continuously read a person, you have to assume that they have moral standards and that sometimes you'll agree with them and sometimes you won't. Morality accepts hierarchies, nuances, and even obsessions. I know not everyone shares my opinions, but at least I can voice them. One day, one of those firms may sue me (as record companies did some years ago).

Palantir is incredibly harmful. Limit its operations. Like Meta and TikTok, its business strategy is shaped by its founders' immorality. Such a procedure can never be beneficial.