VeeFriends is one NFT project I'm sure will last.

I believe in blockchain technology and JPEGs, aka NFTs. NFTs aren't JPEGs. It's not as it seems.

Gary Vaynerchuk is leading the pack with his new NFT project VeeFriends, I wrote a year ago. I was spot-on. It's the most innovative project I've seen.

Since its minting in May 2021, it has given its holders enormous value, most notably the first edition of VeeCon, a multi-day superconference featuring iconic and emerging leaders in NFTs and Popular Culture. First-of-its-kind NFT-ticketed Web3 conference to build friendships, share ideas, and learn together.

VeeFriends holders got free VeeCon NFT tickets. Attendees heard iconic keynote speeches, innovative talks, panels, and Q&A sessions.

It was a unique conference that most of us, including me, are looking forward to in 2023. The lineup was epic, and it allowed many to network in new ways. Really memorable learning. Here are a couple of gratitude posts from the attendees.

VeeFriends Series 2

This article explains VeeFriends if you're still confused.

GaryVee's hand-drawn doodles have evolved into wonderful characters. The characters' poses and backgrounds bring the VeeFriends IP to life.

Yes, this is the second edition of VeeFriends, and at current prices, it's one of the best NFT opportunities in years. If you have the funds and risk appetite to invest in NFTs, VeeFriends Series 2 is worth every penny. Even if you can't invest, learn from their journey.

1. Art Is the Start

Many critics say VeeFriends artwork is below average and not by GaryVee. Art is often the key to future success.

Let's look at one of the first Mickey Mouse drawings. No one would have guessed that this would become one of the most beloved animated short film characters. In Walt Before Mickey, Walt Disney's original mouse Mortimer was less refined.

First came a mouse...

These sketches evolved into Steamboat Willie, Disney's first animated short film.

Fred Moore redesigned the character artwork into what we saw in cartoons as kids. Mickey Mouse's history is here.

Looking at how different cartoon characters have evolved and gained popularity over decades, I believe Series 2 characters like Self-Aware Hare, Kind Kudu, and Patient Pig can do the same.

GaryVee captures this journey on the blockchain and lets early supporters become part of history. Time will tell if it rivals Disney, Pokemon, or Star Wars. Gary has been vocal about this vision.

2. VeeFriends is Intellectual Property for the Coming Generations

Most of us grew up watching cartoons, playing with toys, cards, and video games. Our interactions with fictional characters and the stories we hear shape us.

GaryVee is slowly curating an experience for the next generation with animated videos, card games, merchandise, toys, and more.

VeeFriends UNO, a collaboration with Mattel Creations, features 17 VeeFriends characters.

VeeFriends and Zerocool recently released Trading Cards featuring all 268 Series 1 characters and 15 new ones. Another way to build VeeFriends' collectibles brand.

At Veecon, all the characters were collectible toys. Something will soon emerge.

Kids and adults alike enjoy the YouTube channel's animated shorts and VeeFriends Tunes. Here's a song by the holder's Optimistic Otter-loving daughter.

This VeeFriends story is only the beginning. I'm looking forward to animated short film series, coloring books, streetwear, candy, toys, physical collectibles, and other forms of VeeFriends IP.

3. Veefriends will always provide utilities

Smart contracts can be updated at any time and authenticated on a ledger.

VeeFriends Series 2 gives no promise of any utility whatsoever. GaryVee released no project roadmap. In the first few months after launch, many owners of specific characters or scenes received utilities.

Every benefit or perk you receive helps promote the VeeFriends brand.

Recent partnerships are listed below.

• MaryRuth's Multivitamin Gummies

• Productive Puffin holders from VeeFriends x Primitive

• Pickleball Scene & Clown Holders Only

Pickleball & Competitive Clown Exclusive experience, anteater multivitamin gummies, and Puffin x Primitive merch

Considering the price of NFTs, it may not seem like much. It's just the beginning; you never know what the future holds. No other NFT project offers such diverse, ongoing benefits.

4. Garyvee's team is ready

Gary Vaynerchuk's team and record are undisputed. He's a serial entrepreneur and the Chairman & CEO of VaynerX, which includes VaynerMedia, VaynerCommerce, One37pm, and The Sasha Group.

Gary founded VaynerSports, Resy, and Empathy Wines. He's a Candy Digital Board Member, VCR Group Co-Founder, ArtOfficial Co-Founder, and VeeFriends Creator & CEO. Gary was recently named one of Fortune's Top 50 NFT Influencers.

Gary Vayenerchuk aka GaryVee

Gary documents his daily life as a CEO on social media, which has 34 million followers and 272 million monthly views. GaryVee Audio Experience is a top podcast. He's a five-time New York Times best-seller and sought-after speaker.

Gary can observe consumer behavior to predict trends. He understood these trends early and pioneered them.

• 1997 — Realized e-potential commerce's and started winelibrary.com. In five years, he grew his father's wine business from $3M to $60M.

• 2006 — Realized content marketing's potential and started Wine Library on YouTube. TV

• 2009 — Estimated social media's potential (Web2) and invested in Facebook, Twitter, and Tumblr.

• 2014: Ethereum and Bitcoin investments

• 2021 — Believed in NFTs and Web3 enough to launch VeeFriends

GaryVee isn't all of VeeFriends. Andy Krainak, Dave DeRosa, Adam Ripps, Tyler Dowdle, and others work tirelessly to make VeeFriends a success.

GaryVee has said he'll let other businesses fail but not VeeFriends. We're just beginning his 40-year vision.

I have more confidence than ever in a company with a strong foundation and team.

5. Humans die, but characters live forever

What if GaryVee dies or can't work?

A writer's books can immortalize them. As long as their books exist, their words are immortal. Socrates, Hemingway, Aristotle, Twain, Fitzgerald, and others have become immortal.

Everyone knows Vincent Van Gogh's The Starry Night.

We all love reading and watching Peter Parker, Thor, or Jessica Jones. Their behavior inspires us. Stan Lee's message and stories live on despite his death.

GaryVee represents VeeFriends. Creating characters to communicate ensures that the message reaches even those who don't listen.

Gary wants his values and messages to be omnipresent in 268 characters. Messengers die, but their messages live on.

Gary envisions VeeFriends creating timeless stories and experiences. Ten years from now, maybe every kid will sing Patient Pig.

6. I love the intent.

Gary planned to create Workplace Warriors three years ago when he began designing Patient Panda, Accountable Ant, and Empathy elephant. The project stalled. When NFTs came along, he knew.

Gary wanted to create characters with traits he values, such as accountability, empathy, patience, kindness, and self-awareness. He wants future generations to find these traits cool. He hopes one or more of his characters will become pop culture icons.

These emotional skills aren't taught in schools or colleges, but they're crucial for business and life success. I love that someone is teaching this at scale.

In the end, intent matters.

Humans Are Collectors

Buy and collect things to communicate. Since the 1700s. Medieval people formed communities around hidden metals and stones. Many people still collect stamps and coins, and luxury and fashion are multi-trillion dollar industries. We're collectors.

The early 2020s NFTs will be remembered in the future. VeeFriends will define a cultural and technological shift in this era. VeeFriends Series 1 is the original hand-drawn art, but it's expensive. VeeFriends Series 2 is a once-in-a-lifetime opportunity at $1,000.

If you are new to NFTs, check out How to Buy a Non Fungible Token (NFT) For Beginners

This is a non-commercial article. Not financial or legal advice. Information isn't always accurate. Before making important financial decisions, consult a pro or do your own research.

This post is a summary. Read the full article here

If you think NFTs are awful, check out the art market.

The fervor around NFTs has subsided in recent months due to the crypto market crash and the media's short attention span. They were all anyone could talk about earlier this spring. Last semester, when passions were high and field luminaries were discussing "slurp juices," I asked my students and students from over 20 other universities what they thought of NFTs.

According to many, NFTs were either tasteless pyramid schemes or a new way for artists to make money. NFTs contributed to the climate crisis and harmed the environment, but so did air travel, fast fashion, and smartphones. Some students complained that NFTs were cheap, tasteless, algorithmically generated schlock, but others asked how this was different from other art.

I'm not sure what I expected, but the intensity of students' reactions surprised me. They had strong, emotional opinions about a technology I'd always considered administrative. NFTs address ownership and accounting, like most crypto/blockchain projects.

Art markets can be irrational, arbitrary, and subject to the same scams and schemes as any market. And maybe a few shenanigans that are unique to the art world.

The Fairness Question

Fairness, a deflating moral currency, was the general sentiment (the less of it in circulation, the more ardently we clamor for it.) These students, almost all of whom are artists, complained to the mismatch between the quality of the work in some notable NFT collections and the excessive amounts these items were fetching on the market. They can sketch a Bored Ape or Lazy Lion in their sleep. Why should they buy ramen with school loans while certain swindlers get rich?

I understand students. Art markets are unjust. They can be irrational, arbitrary, and governed by chance and circumstance, like any market. And art-world shenanigans.

Almost every mainstream critique leveled against NFTs applies just as easily to art markets

Over 50% of artworks in circulation are fake, say experts. Sincere art collectors and institutions are upset by the prevalence of fake goods on the market. Not everyone. Wealthy people and companies use art as investments. They can use cultural institutions like museums and galleries to increase the value of inherited art collections. People sometimes buy artworks and use family ties or connections to museums or other cultural taste-makers to hype the work in their collection, driving up the price and allowing them to sell for a profit. Money launderers can disguise capital flows by using market whims, hype, and fluctuating asset prices.

Almost every mainstream critique leveled against NFTs applies just as easily to art markets.

Art has always been this way. Edward Kienholz's 1989 print series satirized art markets. He stamped 395 identical pieces of paper from $1 to $395. Each piece was initially priced as indicated. Kienholz was joking about a strange feature of art markets: once the last print in a series sells for $395, all previous works are worth at least that much. The entire series is valued at its highest auction price. I don't know what a Kienholz print sells for today (inquire with the gallery), but it's more than $395.

I love Lee Lozano's 1969 "Real Money Piece." Lozano put cash in various denominations in a jar in her apartment and gave it to visitors. She wrote, "Offer guests coffee, diet pepsi, bourbon, half-and-half, ice water, grass, and money." "Offer real money as candy."

Lee Lozano kept track of who she gave money to, how much they took, if any, and how they reacted to the offer of free money without explanation. Diverse reactions. Some found it funny, others found it strange, and others didn't care. Lozano rarely says:

Apr 17 Keith Sonnier refused, later screws lid very tightly back on. Apr 27 Kaltenbach takes all the money out of the jar when I offer it, examines all the money & puts it all back in jar. Says he doesn’t need money now. Apr 28 David Parson refused, laughing. May 1 Warren C. Ingersoll refused. He got very upset about my “attitude towards money.” May 4 Keith Sonnier refused, but said he would take money if he needed it which he might in the near future. May 7 Dick Anderson barely glances at the money when I stick it under his nose and says “Oh no thanks, I intend to earn it on my own.” May 8 Billy Bryant Copley didn’t take any but then it was sort of spoiled because I had told him about this piece on the phone & he had time to think about it he said.

Smart Contracts (smart as in fair, not smart as in Blockchain)

Cornell University's Cheryl Finley has done a lot of research on secondary art markets. I first learned about her research when I met her at the University of Florida's Harn Museum, where she spoke about smart contracts (smart as in fair, not smart as in Blockchain) and new protocols that could help artists who are often left out of the economic benefits of their own work, including women and women of color.

Her talk included findings from her ArtNet op-ed with Lauren van Haaften-Schick, Christian Reeder, and Amy Whitaker.

NFTs allow us to think about and hack on formal contractual relationships outside a system of laws that is currently not set up to service our community.

The ArtNet article The Recent Sale of Amy Sherald's ‘Welfare Queen' Symbolizes the Urgent Need for Resale Royalties and Economic Equity for Artists discussed Sherald's 2012 portrait of a regal woman in a purple dress wearing a sparkling crown and elegant set of pearls against a vibrant red background.

Amy Sherald sold "Welfare Queen" to Princeton professor Imani Perry. Sherald agreed to a payment plan to accommodate Perry's budget.

Amy Sherald rose to fame for her 2016 portrait of Michelle Obama and her full-length portrait of Breonna Taylor, one of the most famous works of the past decade.

As is common, Sherald's rising star drove up the price of her earlier works. Perry's "Welfare Queen" sold for $3.9 million in 2021.

Imani Perry's early investment paid off big-time. Amy Sherald, whose work directly increased the painting's value and who was on an artist's shoestring budget when she agreed to sell "Welfare Queen" in 2012, did not see any of the 2021 auction money. Perry and the auction house got that money.

Sherald sold her Breonna Taylor portrait to the Smithsonian and Louisville's Speed Art Museum to fund a $1 million scholarship. This is a great example of what an artist can do for the community if they can amass wealth through their work.

NFTs haven't solved all of the art market's problems — fakes, money laundering, market manipulation — but they didn't create them. Blockchain and NFTs are credited with making these issues more transparent. More ideas emerge daily about what a smart contract should do for artists.

NFTs are a copyright solution. They allow us to hack formal contractual relationships outside a law system that doesn't serve our community.

Amy Sherald shows the good smart contracts can do (as in, well-considered, self-determined contracts, not necessarily blockchain contracts.) Giving back to our community, deciding where and how our work can be sold or displayed, and ensuring artists share in the equity of our work and the economy our labor creates.

10 biases in 10 visuals

Cognitive biases are crucial for UX research, design, and daily life. Our biases distort reality.

After learning about biases at my UX Research bootcamp, I studied Erika Hall's Just Enough Research and used the Nielsen Norman Group's wealth of information. 10 images show my findings.

1. Bias in sampling

Misselection of target population members causes sampling bias. For example, you are building an app to help people with food intolerances log their meals and are targeting adult males (years 20-30), adult females (ages 20-30), and teenage males and females (ages 15-19) with food intolerances. However, a sample of only adult males and teenage females is biased and unrepresentative.

2. Sponsor Disparity

Sponsor bias occurs when a study's findings favor an organization's goals. Beware if X organization promises to drive you to their HQ, compensate you for your time, provide food, beverages, discounts, and warmth. Participants may endeavor to be neutral, but incentives and prizes may bias their evaluations and responses in favor of X organization.

In Just Enough Research, Erika Hall suggests describing the company's aims without naming it.

Third, False-Consensus Bias

False-consensus bias is when a person thinks others think and act the same way. For instance, if a start-up designs an app without researching end users' needs, it could fail since end users may have different wants. https://www.nngroup.com/videos/false-consensus-effect/

Working directly with the end user and employing many research methodologies to improve validity helps lessen this prejudice. When analyzing data, triangulation can boost believability.

Bias of the interviewer

I struggled with this bias during my UX research bootcamp interviews. Interviewing neutrally takes practice and patience. Avoid leading questions that structure the story since the interviewee must interpret them. Nodding or smiling throughout the interview may subconsciously influence the interviewee's responses.

The Curse of Knowledge

The curse of knowledge occurs when someone expects others understand a subject as well as they do. UX research interviews and surveys should reduce this bias because technical language might confuse participants and harm the research. Interviewing participants as though you are new to the topic may help them expand on their replies without being influenced by the researcher's knowledge.

Confirmation Bias

Most prevalent bias. People highlight evidence that supports their ideas and ignore data that doesn't. The echo chamber of social media creates polarization by promoting similar perspectives.

A researcher with confirmation bias may dismiss data that contradicts their research goals. Thus, the research or product may not serve end users.

Design biases

UX Research design bias pertains to study construction and execution. Design bias occurs when data is excluded or magnified based on human aims, assumptions, and preferences.

The Hawthorne Impact

Remember when you behaved differently while the teacher wasn't looking? When you behaved differently without your parents watching? A UX research study's Hawthorne Effect occurs when people modify their behavior because you're watching. To escape judgment, participants may act and speak differently.

To avoid this, researchers should blend into the background and urge subjects to act alone.

The bias against social desire

People want to belong to escape rejection and hatred. Research interviewees may mislead or slant their answers to avoid embarrassment. Researchers should encourage honesty and confidentiality in studies to address this. Observational research may reduce bias better than interviews because participants behave more organically.

Relative Time Bias

Humans tend to appreciate recent experiences more. Consider school. Say you failed a recent exam but did well in the previous 7 exams. Instead, you may vividly recall the last terrible exam outcome.

If a UX researcher relies their conclusions on the most recent findings instead of all the data and results, recency bias might occur.

I hope you liked learning about UX design, research, and real-world biases.

Looking back into the past can help you understand what's happening today and why.

"Creator economy" conjures up images of originality, sincerity, and passion. Where do Michelangelos and da Vincis push advancement with their gifts without battling for bread and proving themselves posthumously? 

Creativity has been as long as humanity, but it's just recently become a new economic paradigm. We even talk about Web3 now.

Let's examine the creative economy's history to better comprehend it. What brought us here? Looking back can help you understand what's happening now.

No yawning, I promise 😉.

Creator Economy's history

Long, uneven transition to creator economy. Let's examine the economic and societal changes that led us there.

1. Agriculture to industry

Mid-18th-century Industrial Revolution led to shift from agriculture to manufacturing. The industrial economy lasted until World War II.

The industrial economy's principal goal was to provide more affordable, accessible commodities.

Unlike today, products were scarce and inaccessible.

To fulfill its goals, industrialization triggered enormous economic changes, moving power from agrarians to manufacturers. Industrialization brought hard work, rivalry, and new ideas connected to production and automation. Creative thinkers focused on that then.

It doesn't mean music, poetry, or painting had no place back then. They weren't top priority. Artists were independent. The creative field wasn't considered a different economic subdivision.

2. The consumer economy

Manufacturers produced more things than consumers desired after World War II. Stuff was no longer scarce.

The economy must make customers want to buy what the market offers.

The consumer economic paradigm supplanted the industrial one. Customers (or consumers) replaced producers as the new economic center.

Salesmen, marketing, and journalists also played key roles (TV, radio, newspapers, etc.). Mass media greatly boosted demand for goods, defined trends, and changed views regarding nearly everything.

Mass media also gave rise to pop culture, which focuses on mass-market creative products. Design, printing, publishing, multi-media, audio-visual, cinematographic productions, etc. supported pop culture.

The consumer paradigm generated creative occupations and activities, unlike the industrial economy. Creativity was limited by the need for wide appeal.

Most creators were corporate employees.

Creating a following and making a living from it were difficult.

Paul Saffo said that only journalists and TV workers were known. Creators who wished to be known relied on producers, publishers, and other gatekeepers. To win their favor was crucial. Luck was the best tactic.

3. The creative economy

Consumer economy was digitized in the 1990s. IT solutions transformed several economic segments. This new digital economy demanded innovative, digital creativity.

Later, states declared innovation a "valuable asset that creates money and jobs." They also introduced the "creative industries" and the "creative economy" (not creator!) and tasked themselves with supporting them. Australia and the UK were early adopters.

Individual skill, innovation, and intellectual property fueled the creative economy. Its span covered design, writing, audio, video material, etc. The creative economy required IT-powered activity.

The new challenge was to introduce innovations to most economic segments and meet demand for digital products and services.

Despite what the title "creative economy" may imply, it was primarily oriented at meeting consumer needs. It didn't provide inventors any new options to become entrepreneurs. Instead of encouraging innovators to flourish on their own, the creative economy emphasized "employment-based creativity."

4. The creator economy

Next, huge IT platforms like Google, Facebook, YouTube, and others competed with traditional mainstream media.

During the 2008 global financial crisis, these mediums surpassed traditional media. People relied on them for information, knowledge, and networking. That was a digital media revolution. The creator economy started there.

The new economic paradigm aimed to engage and convert clients. The creator economy allowed customers to engage, interact, and provide value, unlike the consumer economy. It gave them instruments to promote themselves as "products" and make money.

Writers, singers, painters, and other creators have a great way to reach fans. Instead of appeasing old-fashioned gatekeepers (producers, casting managers, publishers, etc.), they can use the platforms to express their talent and gain admirers. Barriers fell.

It's not only for pros. Everyone with a laptop and internet can now create.

2022 creator economy:

Since there is no academic description for the current creator economy, we can freestyle.

The current (or Web2) creator economy is fueled by interactive digital platforms, marketplaces, and tools that allow users to access, produce, and monetize content.

No entry hurdles or casting in the creative economy. Sign up and follow platforms' rules. Trick: A platform's algorithm aggregates your data and tracks you. This is the payment for participation.

The platforms offer content creation, design, and ad distribution options. This is platforms' main revenue source.

The creator economy opens many avenues for creators to monetize their work. Artists can now earn money through advertising, tipping, brand sponsorship, affiliate links, streaming, and other digital marketing activities.

Even if your content isn't digital, you can utilize platforms to promote it, interact and convert your audience, and more. No limits. However, some of your income always goes to a platform (well, a huge one).

The creator economy aims to empower online entrepreneurship by offering digital marketing tools and reducing impediments.

Barriers remain. They are just different. Next articles will examine these.

Why update the creator economy for Web3?

I could address this question by listing the present creator economy's difficulties that led us to contemplate a Web3 upgrade.

I don't think these difficulties are the main cause. The mentality shift made us see these challenges and understand there was a better reality without them.

Crypto drove this thinking shift. It promoted disintermediation, independence from third-party service providers, 100% data ownership, and self-sovereignty. Crypto has changed the way we view everyday things.

Crypto's disruptive mission has migrated to other economic segments. It's now called Web3. Web3's creator economy is unique.

Here's the essence of the Web3 economy:

• Eliminating middlemen between creators and fans.

• 100% of creators' data, brand, and effort.

• Business and money-making transparency.

• Authentic originality above ad-driven content.

In the next several articles, I'll explain. We'll also discuss the creator economy and Web3's remedies.

Final thoughts

The creator economy is the organic developmental stage we've reached after all these social and economic transformations.

The Web3 paradigm of the creator economy intends to allow creators to construct their own independent "open economy" and directly monetize it without a third party.

If this approach succeeds, we may enter a new era of wealth creation where producers aren't only the products. New economies will emerge.

This article is a summary. To read the full post, click here.

The Federal Funds Rate (FFR) is the target interest rate set by the Federal Reserve System (Fed)'s policy-making body (FOMC). This target is the rate at which the Fed suggests commercial banks borrow and lend their excess reserves overnight to each other.

The FOMC meets 8 times a year to set the target FFR. This is supposed to promote economic growth. The overnight lending market sets the actual rate based on commercial banks' short-term reserves. If the market strays too far, the Fed intervenes.

Banks must keep a certain percentage of their deposits in a Federal Reserve account. A bank's reserve requirement is a percentage of its total deposits. End-of-day bank account balances averaged over two-week reserve maintenance periods are used to determine reserve requirements.

If a bank expects to have end-of-day balances above what's needed, it can lend the excess to another institution.

The FOMC adjusts interest rates based on economic indicators that show inflation, recession, or other issues that affect economic growth. Core inflation and durable goods orders are indicators.

In response to economic conditions, the FFR target has changed over time. In the early 1980s, inflation pushed it to 20%. During the Great Recession of 2007-2009, the rate was slashed to 0.15 percent to encourage growth.

Inflation picked up in May 2022 despite earlier rate hikes, prompting today's 0.75 percent point increase. The largest increase since 1994. It might rise to around 3.375% this year and 3.1% by the end of 2024.

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.