The Jordan 6 Rings Reintroduce Classic Bulls
The Jordan 6 Rings return in Bulls colors, a deviation from previous releases. The signature red color is used on the midsole and heel, as well as the chenille patch and pull tab. The rest of the latter fixture is black, matching the outsole and adjacent Jumpman logos. Finally, white completes the look, from the leather mudguard to the lace unit. Here's a closer look at the Jordan 6 Rings. Sizes should be available soon on Nike.com and select retailers. Also, official photos of the Air Jordan 1 Denim have surfaced.
Jordan 6 Rings
Release Date: 2022
Color: N/A
Mens: $130
Style Code: 322992-126
More on Lifestyle
Josh Chesler
3 years ago
10 Sneaker Terms Every Beginner Should Know
So you want to get into sneakers? Buying a few sneakers and figuring it out seems simple. Then you miss out on the weekend's instant-sellout releases, so you head to eBay, Twitter, or your local sneaker group to see what's available, since you're probably not ready to pay Flight Club prices just yet.
That's when you're bombarded with new nicknames, abbreviations, and general sneaker slang. It would take months to explain every word and sneaker, so here's a starter kit of ten simple terms to get you started. (Yeah, mostly Jordan. Does anyone really start with Kith or Nike SB?)
10. Colorways
Colorways are a common term in fashion, design, and other visual fields. It's just the product's color scheme. In the case of sneakers, the colorway is often as important as the actual model. Are this year's "Chicago" Air Jordan 1s more durable than last year's "Black/Gum" colorway? Because of their colorway and rarity, the Chicagos are worth roughly three pairs of the Black/Gum kicks.
Pro Tip: A colorway with a well-known nickname is almost always worth more than one without, and the same goes for collaborations.
9. Beaters
A “beater” is a well-worn, likely older model of shoe that has significant wear and tear on it. Rarely sold with the original box or extra laces, beaters rarely sell for much. Unlike most “worn” sneakers, beaters are used for rainy days and the gym. It's exactly what it sounds like, a box full of beaters, and they're a good place to start if you're looking for some cheap old kicks.
Pro Tip: Know which shoes clean up nicely. The shape of lower top sneakers with wider profiles, like SB Dunk Lows and Air Jordan 3s, tends to hold better over time than their higher and narrower cousins.
8. Retro
In the world of Jordan Brand, a “Retro” release is simply a release (or re-release) of a colorway after the shoe model's initial release. For example, the original Air Jordan 7 was released in 1992, but the Bordeaux colorway was re-released in 2011 and recently (2015). An Air Jordan model is released every year, and while half of them are unpopular and unlikely to be Retroed soon, any of them could be re-released whenever Nike and Jordan felt like it.
Pro Tip: Now that the Air Jordan line has been around for so long, the model that tends to be heavily retroed in a year is whichever shoe came out 23 (Michael Jordan’s number during the prime of his career) years ago. The Air Jordan 6 (1991) got new colorways last year, the Air Jordan 7 this year, and more Air Jordan 8s will be released later this year and early next year (1993).
7. PP/Inv
In spite of the fact that eBay takes roughly 10% of the final price, many sneaker buyers and sellers prefer to work directly with PayPal. Selling sneakers for $100 via PayPal invoice or $100 via PayPal friends/family is common on social media. Because no one wants their eBay account suspended for promoting PayPal deals, many eBay sellers will simply state “Message me for a better price.”
Pro Tip: PayPal invoices protect buyers well, but gifting or using Google Wallet does not. Unless you're certain the seller is legitimate, only use invoiced goods/services payments.
6. Yeezy
Kanye West and his sneakers are known as Yeezys. The rapper's first two Yeezys were made by Nike before switching to Adidas. Everything Yeezy-related will be significantly more expensive (and therefore have significantly more fakes made). Not only is the Nike Air Yeezy 2 “Red October” one of the most sought-after sneakers, but the Yeezy influence can be seen everywhere.
Pro Tip: If you're going to buy Yeezys, make sure you buy them from a reputable retailer or reseller. With so many fakes out there, it's not worth spending a grand on something you're not 100% sure is real.
5. GR/Limited
Regardless of how visually repulsive, uncomfortable, and/or impractical a sneaker is, if it’s rare enough, people will still want it. GR stands for General Release, which means they're usually available at retail. Reselling a “Limited Edition” release is costly. Supply and demand, but in this case, the limited supply drives up demand. If you want to get some of the colorways made for rappers, NBA players (Player Exclusive or PE models), and other celebrities, be prepared to pay a premium.
Pro Tip: Limited edition sneakers, like the annual Doernbecher Freestyle sneakers Nike creates with kids from Portland's Doernbecher Children's Hospital, will always be more expensive and limited. Or, you can use automated sneaker-buying software.
4. Grails
A “grail” is a pair of sneakers that someone desires above all others. To obtain their personal grails, people are willing to pay significantly more than the retail price. There doesn't have to be any rhyme or reason why someone chose a specific pair as their grails.
Pro Tip: For those who don't have them, the OG "Bred" or "Royal" Air Jordan 1s, the "Concord" Air Jordan 11s, etc., are all grails.
3. Bred
Anything released in “Bred” (black and red) will sell out quickly. Most resale Air Jordans (and other sneakers) come in the Bred colorway, which is a fan favorite. Bred is a good choice for a first colorway, especially on a solid sneaker silhouette.
Pro Tip: Apart from satisfying the world's hypebeasts, Bred sneakers will probably match a lot of your closet.
2. DS
DS = Deadstock = New. That's it. If something has been worn or tried on, it is no longer DS. Very Near Deadstock (VNDS) Pass As Deadstock It's a cute way of saying your sneakers have been worn but are still in good shape. In the sneaker world, “worn” means they are no longer new, but not too old or beat up.
Pro Tip: Ask for photos of any marks or defects to see what you’re getting before you buy used shoes, also find out if they come with the original box and extra laces, because that can be a sign that they’re in better shape.
1. Fake/Unauthorized
The words “Unauthorized,” “Replica,” “B-grades,” and “Super Perfect” all mean the shoes are fake. It means they aren't made by the actual company, no matter how close or how good the quality. If that's what you want, go ahead and get them. Do not wear them if you do not want the rest of the sneaker world to mock them.
Pro Tip: If you’re not sure if shoes are real or not, do a “Legit Check” on Twitter or Facebook. You'll get dozens of responses in no time.
Hannah Elliott
2 years ago
Pebble Beach Auto Auctions Set $469M Record
The world's most prestigious vintage vehicle show included amazing autos and record-breaking sums.
This 1932 Duesenberg J Figoni Sports Torpedo earned Best of Show in 2022.
David Paul Morris (DPM)/Bloomberg
2022 Pebble Beach Concours d'Elegance winner was a pre-war roadster.
Lee Anderson's 1932 Duesenberg J Figoni Sports Torpedo won Best of Show at Pebble Beach Golf Links near Carmel, Calif., on Sunday. First American win since 2013.
Sandra Button, chairperson of the annual concours, said the car, whose chassis and body had been separated for years, "marries American force with European style." "Its resurrection story is passionate."
Pebble Beach Concours d'Elegance Auction
Since 1950, the Pebble Beach Concours d'Elegance has welcomed the world's most costly collectable vehicles for a week of parties, auctions, rallies, and high-roller meetings. The cold, dreary weather highlighted the automobiles' stunning lines and hues.
DPM/Bloomberg
A visitor photographs a 1948 Ferrari 166 MM Touring Barchetta. This is one of 25 Ferraris manufactured in the years after World War II. First shown at the 1948 Turin Salon. Others finished Mille Miglia and Le Mans, which set the tone for Ferrari racing for years.
DPM/Bloomberg
This year's frontrunners were ultra-rare pre-war and post-war automobiles with long and difficult titles, such a 1937 Talbot-Lago T150C-SS Figoni & Falaschi Teardrop Coupe and a 1951 Talbot-Lago T26 Grand Sport Stabilimenti Farina Cabriolet.
The hefty, enormous coaches inspire visions of golden pasts when mysterious saloons swept over the road with otherworldly style, speed, and grace. Only the richest and most powerful people, like Indian maharaja and Hollywood stars, owned such vehicles.
Antonio Chopitea, a Peruvian sugar tycoon, ordered a new Duesenberg in Paris. Hemmings says the two-tone blue beauty was moved to the US and dismantled in the 1960s. Body and chassis were sold separately and rejoined decades later in a three-year, prize-winning restoration.
The concours is the highlight of Monterey Car Week, a five-day Super Bowl for car enthusiasts. Early events included Porsche and Ferrari displays, antique automobile races, and new-vehicle debuts. Many auto executives call Monterey Car Week the "new auto show."
Many visitors were drawn to the record-breaking auctions.
A 1969 Porsche 908/02 auctioned for $4.185 million. Flat-eight air-cooled engine, 90.6-inch wheelbase, 1,320-pound weight. Vic Elford, Richard Attwood, Rudi Lins, Gérard Larrousse, Kurt Ahrens Jr., Masten Gregory, and Pedro Rodriguez drove it, according to Gooding.
DPM/Bloomberg
The 1931 Bentley Eight Liter Sports Tourer doesn't meet its reserve. Gooding & Co., the official auction house of the concours, made more than $105 million and had an 82% sell-through rate. This powerful open-top tourer is one of W.O. Bentley's 100 automobiles. Only 80 remain.
DPM/Bloomberg
The final auction on Aug. 21 brought in $456.1 million, breaking the previous high of $394.48 million established in 2015 in Monterey. “The week put an exclamation point on what has been an exceptional year for the collector automobile market,” Hagerty analyst John Wiley said.
Many cars that go unsold at public auction are sold privately in the days after. After-sales pushed the week's haul to $469 million on Aug. 22, up 18.9% from 2015's record.
In today's currencies, 2015's record sales amount to $490 million, Wiley noted. The dollar is degrading faster than old autos.
Still, 113 million-dollar automobiles sold. The average car sale price was $583,211, up from $446,042 last year, while multimillion-dollar hammer prices made up around 75% of total sales.
Industry insiders and market gurus expected that stock market volatility, the crisis in Ukraine, and the dollar-euro exchange rate wouldn't influence the world's biggest spenders.
Classic.com's CEO said there's no hint of a recession in an e-mail. Big sales and crowds.
Ticket-holders wore huge hats, flowery skirts, and other Kentucky Derby-esque attire. Coffee, beverages, and food are extra.
DPM/Bloomberg
Mercedes-Benz 300 SL Gullwing, 1955. Mercedes produced the two-seat gullwing coupe from 1954–1957 and the roadster from 1957–1963. It was once West Germany's fastest and most powerful automobile. You'd be hard-pressed to locate one for less $1 million.
DPM/Bloomberg
1955 Ferrari 410 Sport sold for $22 million at RM Sotheby's. It sold a 1937 Mercedes-Benz 540K Sindelfingen Roadster for $9.9 million and a 1924 Hispano-Suiza H6C Transformable Torpedo for $9.245 million. The family-run mansion sold $221.7 million with a 90% sell-through rate, up from $147 million in 2021. This year, RM Sotheby's cars averaged $1.3 million.
Not everyone saw such great benefits.
Gooding & Co., the official auction house of the concours, made more than $105 million and had an 82% sell-through rate. 1937 Bugatti Type 57SC Atalante, 1990 Ferrari F40, and 1994 Bugatti EB110 Super Sport were top sellers.
The 1969 Autobianchi A112 Bertone. This idea two-seater became a Hot Wheels toy but was never produced. It has a four-speed manual drive and an inline-four mid-engine arrangement like the Lamborghini Miura.
DPM/Bloomberg
1956 Porsche 356 A Speedster at Gooding & Co. The Porsche 356 is a lightweight, rear-engine, rear-wheel drive vehicle that lacks driving power but is loved for its rounded, Beetle-like hardtop coupé and open-top versions.
DPM/Bloomberg
Mecum sold $50.8 million with a 64% sell-through rate, down from $53.8 million and 77% in 2021. Its top lot, a 1958 Ferrari 250 GT 'Tour de France' Alloy Coupe, sold for $2.86 million, but its average price was $174,016.
Bonhams had $27.8 million in sales with an 88% sell-through rate. The same sell-through generated $35.9 million in 2021.
Gooding & Co. and RM Sotheby's posted all 10 top sales, leaving Bonhams, Mecum, and Hagerty-owned Broad Arrow fighting for leftovers. Six of the top 10 sellers were Ferraris, which remain the gold standard for collectable automobiles. Their prices have grown over decades.
Classic.com's Calle claimed RM Sotheby's "stole the show," but "BroadArrow will be a force to reckon with."
Although pre-war cars were hot, '80s and '90s cars showed the most appreciation and attention. Generational transition and new buyer profile."
2022 Pebble Beach Concours d'Elegance judges inspect 1953 Siata 208. The rounded coupe was introduced at the 1952 Turin Auto Show in Italy and is one of 18 ever produced. It sports a 120hp Fiat engine, five-speed manual transmission, and alloy drum brakes. Owners liked their style, but not their reliability.
DPM/Bloomberg
The Czinger 21 CV Max at Pebble Beach. Monterey Car Week concentrates on historic and classic automobiles, but modern versions like this Czinger hypercar also showed.
DPM/Bloomberg
The 1932 Duesenberg J Figoni Sports Torpedo won Best in Show in 2022. Lee and Penny Anderson of Naples, Fla., own the once-separate-chassis-from-body automobile.
DPM/Bloomberg
Marcus Lu
2 years ago
The Brand Structure of U.S. Electric Vehicle Production
Will Tesla be able to maintain its lead in the EV market for very long?
This is one of the most pressing issues in the American auto sector today. One positive aspect of Tesla is the company's devoted customer base and recognizable name recognition (similar to Apple). It also invests more in research and development per vehicle than its rivals and has a head start in EV production.
Conversely, established automakers like Volkswagen are actively plotting their strategy to surpass Tesla. As the current market leaders, they have decades of experience in the auto industry and are spending billions to catch up.
We've visualized data from the EPA's 2022 Automotive Trends Report to bring you up to speed on this developing story.
Info for the Model Year of 2021
The full production data used in this infographic is for the 2021 model year, but it comes from a report for 2022.
Combined EV and PHEV output is shown in the table below (plug-in hybrid electric vehicle).
It is important to note that Toyota and Stellantis, the two largest legacy automakers in this dataset, only produced PHEVs. Toyota's first electric vehicle, the bZ4X, won't hit the market until 2023.
Stellantis seems to be falling even further behind, despite having enormous unrealized potential in its Jeep and Ram brands. Stellantis CEO Carlos Tavares said in a recent interview that the firm has budgeted $36 billion for electrification and software.
Legacy Brands with the Most Momentum
In the race to develop electric vehicles, some long-standing manufacturers have gotten the jump on their rivals.
Volkswagen, one of these storied manufacturers, has made a significant investment in electric vehicles (EVs) in the wake of the Dieselgate scandal. The company plans to roll out multiple EV models, including the ID.3 hatchback, ID.4 SUV, and ID. Buzz, with the goal of producing 22 million EVs by 2028. (an electric revival of the classic Microbus).
Even Ford is keeping up, having just announced an EV investment of $22 billion between 2021 and 2025. In November of 2022, the company manufactured their 150,000th Mustang Mach-E, and by the end of 2023, they hoped to have 270,000 of them in circulation.
Additionally, over 200,000 F-150 Lightnings have been reserved since Ford announced the truck. The Lightning is scheduled to have a production run of 15,000 in 2022, 55,000 in 2023, and 80,000 in 2024. Ford's main competitor in the electric pickup truck segment, Rivian, is on track to sell 25,000 vehicles by 2022.
You might also like

Ian Writes
2 years ago
Rich Dad, Poor Dad is a Giant Steaming Pile of Sh*t by Robert Kiyosaki.
Don't promote it.
I rarely read a post on how Rich Dad, Poor Dad motivated someone to grow rich or change their investing/finance attitude. Rich Dad, Poor Dad is a sham, though. This book isn't worth anyone's attention.
Robert Kiyosaki, the author of this garbage, doesn't deserve recognition or attention. This first finance guru wanted to build his own wealth at your expense. These charlatans only care about themselves.
The reason why Rich Dad, Poor Dad is a huge steaming piece of trash
The book's ideas are superficial, apparent, and unsurprising to entrepreneurs and investors. The book's themes may seem profound to first-time readers.
Apparently, starting a business will make you rich.
The book supports founding or buying a business, making it self-sufficient, and being rich through it. Starting a business is time-consuming, tough, and expensive. Entrepreneurship isn't for everyone. Rarely do enterprises succeed.
Robert says we should think like his mentor, a rich parent. Robert never said who or if this guy existed. He was apparently his own father. Robert proposes investing someone else's money in several enterprises and properties. The book proposes investing in:
“have returns of 100 percent to infinity. Investments that for $5,000 are soon turned into $1 million or more.”
In rare cases, a business may provide 200x returns, but 65% of US businesses fail within 10 years. Australia's first-year business failure rate is 60%. A business that lasts 10 years doesn't mean its owner is rich. These statistics only include businesses that survive and pay their owners.
Employees are depressed and broke.
The novel portrays employees as broke and sad. The author degrades workers.
I've owned and worked for a business. I was broke and miserable as a business owner, working 80 hours a week for absolutely little salary. I work 50 hours a week and make over $200,000 a year. My work is hard, intriguing, and I'm surrounded by educated individuals. Self-employed or employee?
Don't listen to a charlatan's tax advice.
From a bad advise perspective, Robert's tax methods were funny. Robert suggests forming a corporation to write off holidays as board meetings or health club costs as business expenses. These actions can land you in serious tax trouble.
Robert dismisses college and traditional schooling. Rich individuals learn by doing or living, while educated people are agitated and destitute, says Robert.
Rich dad says:
“All too often business schools train employees to become sophisticated bean-counters. Heaven forbid a bean counter takes over a business. All they do is look at the numbers, fire people, and kill the business.”
And then says:
“Accounting is possibly the most confusing, boring subject in the world, but if you want to be rich long-term, it could be the most important subject.”
Get rich by avoiding paying your debts to others.
While this book has plenty of bad advice, I'll end with this: Robert advocates paying yourself first. This man's work with Trump isn't surprising.
Rich Dad's book says:
“So you see, after paying myself, the pressure to pay my taxes and the other creditors is so great that it forces me to seek other forms of income. The pressure to pay becomes my motivation. I’ve worked extra jobs, started other companies, traded in the stock market, anything just to make sure those guys don’t start yelling at me […] If I had paid myself last, I would have felt no pressure, but I’d be broke.“
Paying yourself first shouldn't mean ignoring debt, damaging your credit score and reputation, or paying unneeded fees and interest. Good business owners pay employees, creditors, and other costs first. You can pay yourself after everyone else.
If you follow Robert Kiyosaki's financial and business advice, you might as well follow Donald Trump's, the most notoriously ineffective businessman and swindle artist.
This book's popularity is unfortunate. Robert utilized the book's fame to promote paid seminars. At these seminars, he sold more expensive seminars to the gullible. This strategy was utilized by several conmen and Trump University.
It's reasonable that many believed him. It sounded appealing because he was pushing to get rich by thinking like a rich person. Anyway. At a time when most persons addressing wealth development advised early sacrifices (such as eschewing luxury or buying expensive properties), Robert told people to act affluent now and utilize other people's money to construct their fantasy lifestyle. It's exciting and fast.
I often voice my skepticism and scorn for internet gurus now that social media and platforms like Medium make it easier to promote them. Robert Kiyosaki was a guru. Many people still preach his stuff because he was so good at pushing it.

Stephen Moore
2 years ago
Adam Neumanns is working to create the future of living in a classic example of a guy failing upward.
The comeback tour continues…
First, he founded a $47 billion co-working company (sorry, a “tech company”).
He established WeLive to disrupt apartment life.
Then he created WeGrow, a school that tossed aside the usual curriculum to feed children's souls and release their potential.
He raised the world’s consciousness.
Then he blew it all up (without raising the world’s consciousness). (He bought a wave pool.)
Adam Neumann's WeWork business burned investors' money. The founder sailed off with unimaginable riches, leaving long-time employees with worthless stocks and the company bleeding money. His track record, which includes a failing baby clothing company, should have stopped investors cold.
Once the dust settled, folks went on. We forgot about the Neumanns! We forgot about the private jets, company retreats, many houses, and WeWork's crippling. In that moment, the prodigal son of entrepreneurship returned, choosing the blockchain as his industry. His homecoming tour began with Flowcarbon, which sold Goddess Nature Tokens to lessen companies' carbon footprints.
Did it work?
Of course not.
Despite receiving $70 million from Andreessen Horowitz's a16z, the project has been halted just two months after its announcement.
This triumph should lower his grade.
Neumann seems to have moved on and has another revolutionary idea for the future of living. Flow (not Flowcarbon) aims to help people live in flow and will launch in 2023. It's the classic Neumann pitch: lofty goals, yogababble, and charisma to attract investors.
It's a winning formula for one investment fund. a16z has backed the project with its largest single check, $350 million. It has a splash page and 3,000 rental units, but is valued at over $1 billion. The blog post praised Neumann for reimagining the office and leading a paradigm-shifting global company.
Flow's mission is to solve the nation's housing crisis. How? Idk. It involves offering community-centric services in apartment properties to the same remote workforce he once wooed with free beer and a pingpong table. Revolutionary! It seems the goal is to apply WeWork's goals of transforming physical spaces and building community to apartments to solve many of today's housing problems.
The elevator pitch probably sounded great.
At least a16z knows it's a near-impossible task, calling it a seismic shift. Marc Andreessen opposes affordable housing in his wealthy Silicon Valley town. As details of the project emerge, more investors will likely throw ethics and morals out the window to go with the flow, throwing money at a man known for burning through it while building toxic companies, hoping he can bank another fantasy valuation before it all crashes.
Insanity is repeating the same action and expecting a different result. Everyone on the Neumann hype train needs to sober up.
Like WeWork, this venture Won’tWork.
Like before, it'll cause a shitstorm.

Dmitrii Eliuseev
2 years ago
Creating Images on Your Local PC Using Stable Diffusion AI
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.