As part of Meta's endeavor to establish an interactive live-streaming platform, the company is testing with influencers.

The NPE (new product experimentation team) has been testing Super since late 2020.

Bloomberg defined Super as a Cameo-inspired FaceTime-like gadget in 2020. The tool has evolved into a Twitch-like live streaming application.

Less than 100 creators have utilized Super: Creators can request access on Meta's website. Super isn't an Instagram, Facebook, or Meta extension.

“It’s a standalone project,” the spokesperson said about Super. “Right now, it’s web only. They have been testing it very quietly for about two years. The end goal [of NPE projects] is ultimately creating the next standalone project that could be part of the Meta family of products.” The spokesperson said the outreach this week was part of a drive to get more creators to test Super.

A 2021 pitch deck from Super reveals the inner workings of Meta.

The deck gathered feedback on possible sponsorship models, with mockups of brand deals & features. Meta reportedly paid creators $200 to $3,000 to test Super for 30 minutes.

Meta's pitch deck for Super live streaming was leaked.

What were the slides in the pitch deck for Metas Super?

Embed not supported: see full deck & article here →

View examples of Meta's pitch deck for Super:

Product Slides, first

The pitch deck begins with Super's mission:

Super is a Facebook-incubated platform which helps content creators connect with their fans digitally, and for super fans to meet and support their favorite creators. In the spirit of Late Night talk shows, we feature creators (“Superstars”), who are guests at a live, hosted conversation moderated by a Host.

This slide (and most of the deck) is text-heavy, with few icons, bullets, and illustrations to break up the content. Super's online app status (which requires no download or installation) might be used as a callout (rather than paragraph-form).

Meta's Super platform focuses on brand sponsorships and native placements, as shown in the slide above.

One of our theses is the idea that creators should benefit monetarily from their Super experiences, and we believe that offering a menu of different monetization strategies will enable the right experience for each creator. Our current focus is exploring sponsorship opportunities for creators, to better understand what types of sponsor placements will facilitate the best experience for all Super customers (viewers, creators, and advertisers).

Colorful mockups help bring Metas vision for Super to life.

2. Slide Features

Super's pitch deck focuses on the platform's features. The deck covers pre-show, pre-roll, and post-event for a Sponsored Experience.

• Pre-show: active 30 minutes before the show's start

• Pre-roll: Play a 15-minute commercial for the sponsor before the event (auto-plays once)

• Meet and Greet: This event can have a branding, such as Meet & Greet presented by [Snickers]

• Super Selfies: Makers and followers get a digital souvenir to post on social media.

• Post-Event: Possibility to draw viewers' attention to sponsored content/links during the after-show

Almost every screen displays the Sponsor logo, link, and/or branded background. Viewers can watch sponsor video while waiting for the event to start.

Slide 3: Business Model

Meta's presentation for Super is incomplete without numbers. Super's first slide outlines the creator, sponsor, and Super's obligations. Super does not charge creators any fees or commissions on sponsorship earnings.

How to make a great pitch deck

We hope you can use the Super pitch deck to improve your business. Bestpitchdeck.com/super-meta is a bookmarkable link.

You can also use one of our expert-designed templates to generate a pitch deck.

Our team has helped close $100M+ in agreements and funding for premier companies and VC firms. Use our presentation templates, one-pagers, or financial models to launch your pitch.

Every pitch must be audience-specific. Our team has prepared pitch decks for various sectors and fundraising phases.

Pitch Deck Software VIP.graphics produced a popular SaaS & Software Pitch Deck based on decks that closed millions in transactions & investments for orgs of all sizes, from high-growth startups to Fortune 100 enterprises. This easy-to-customize PowerPoint template includes ready-made features and key slides for your software firm.

Accelerator Pitch Deck The Accelerator Pitch Deck template is for early-stage founders seeking funding from pitch contests, accelerators, incubators, angels, or VC companies. Winning a pitch contest or getting into a top accelerator demands a strategic investor pitch.

Pitch Deck Template Series Startup and founder pitch deck template: Workable, smart slides. This pitch deck template is for companies, entrepreneurs, and founders raising seed or Series A finance.

M&A Pitch Deck Perfect Pitch Deck is a template for later-stage enterprises engaging more sophisticated conversations like M&A, late-stage investment (Series C+), or partnerships & funding. Our team prepared this presentation to help creators confidently pitch to investment banks, PE firms, and hedge funds (and vice versa).

Browse our growing variety of industry-specific pitch decks.

Leonardo—The New Midjourney?

Users are comparing the new cowboy to Midjourney.

Leonardo.AI creates great photographs and has several unique capabilities I haven't seen in other AI image systems.

Midjourney's quality photographs are evident in the community feed.

Create Pictures Using Models

You can make graphics using platform models when you first enter the app (website):

Luma, Leonardo creative, Deliberate 1.1.

Clicking a model displays its description and samples:

Click Generate With This Model.

Then you can add your prompt, alter models, photos, sizes, and guide scale in a sleek UI.

Changing Pictures

Leonardo's Canvas editor lets you change created images by hovering over them:

The editor opens with masking, erasing, and picture download.

Develop Your Own Models

I've never seen anything like Leonardo's model training feature.

Upload a handful of similar photographs and save them as a model for future images. Share your model with the community.

You can make photos using your own model and a community-shared set of fine-tuned models:

Obtain Leonardo access

Leonardo is currently free.

Visit Leonardo.ai and click "Get Early Access" to receive access.

Add your email to receive a link to join the discord channel. Simply describe yourself and fill out a form to join the discord channel.

Please go to 👑│introductions to make an introduction and ✨│priority-early-access will be unlocked, you must fill out a form and in 24 hours or a little more (due to demand), the invitation will be sent to you by email.

I got access in two hours, so hopefully you can too.

Last Words

I know there are many AI generative platforms, some free and some expensive, but Midjourney produces the most artistically stunning images and art.

Leonardo is the closest I've seen to Midjourney, but Midjourney is still the leader.

It's free now.

Leonardo's fine-tuned model selections, model creation, image manipulation, and output speed and quality make it a great AI image toolbox addition.

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.

My wallpape

Traditional sales organizations used the concept of a sales “funnel” to describe the process through which potential customers move, ending up with sales at the end. Winners today have abandoned that way of thinking in favor of building flywheels — business models in which every element reinforces every other.

Ah, the marketing funnel…

Prospective clients go through a predictable set of experiences, students learn in business school marketing classes. It looks like this:

Understanding the funnel helps evaluate sales success indicators. Gail Goodwin, former CEO of small business direct mail provider Constant Contact, said managing the pipeline was key to escaping the sluggish SaaS ramp of death.

Like the funnel concept. To predict how well your business will do, measure how many potential clients are aware of it (awareness) and how many take the next step. If 1,000 people heard about your offering and 10% showed interest, you'd have 100 at that point. If 50% of these people made buyer-like noises, you'd know how many were, etc. It helped model buying trends.

TV, magazine, and radio advertising are pricey for B2C enterprises. Traditional B2B marketing involved armies of sales reps, which was expensive and a barrier to entry.

Cracks in the funnel model

Digital has exposed the funnel's limitations. Hubspot was born at a time when buyers and sellers had huge knowledge asymmetries, according to co-founder Brian Halligan. Those selling a product could use the buyer's lack of information to become a trusted partner.

As the world went digital, getting information and comparing offerings became faster, easier, and cheaper. Buyers didn't need a seller to move through a funnel. Interactions replaced transactions, and the relationship didn't end with a sale.

Instead, buyers and sellers interacted in a constant flow. In many modern models, the sale is midway through the process (particularly true with subscription and software-as-a-service models). Example:

You're creating a winding journey with many touch points, not a funnel (and lots of opportunities for customers to get lost).

From winding journey to flywheel

Beyond this revised view of an interactive customer journey, a company can create what Jim Collins famously called a flywheel. Imagine rolling a heavy disc on its axis. The first few times you roll it, you put in a lot of effort for a small response. The same effort yields faster turns as it gains speed. Over time, the flywheel gains momentum and turns without your help.

Modern digital organizations have created flywheel business models, in which any additional force multiplies throughout the business. The flywheel becomes a force multiplier, according to Collins.

Amazon is a famous flywheel example. Collins explained the concept to Amazon CEO Jeff Bezos at a corporate retreat in 2001. In The Everything Store, Brad Stone describes in his book The Everything Store how he immediately understood Amazon's levers.

The result (drawn on a napkin):

Low prices and a large selection of products attracted customers, while a focus on customer service kept them coming back, increasing traffic. Third-party sellers then increased selection. Low-cost structure supports low-price commitment. It's brilliant! Every wheel turn creates acceleration.

Where from here?

Flywheel over sales funnel! Consider these business terms.

Don’t give a number

Google pays its software engineers generously. While many of their employees are competent, they disregard a critical skill to maximize their pay.

Negotiation.

If Google employees have never negotiated, they're as helpless as anyone else.

In this piece, I'll reveal a compensation negotiation tip that will set you apart.

The Fallacy of Negotiating

How do you negotiate your salary? “Just give them a number twice the amount you really want”. - Someplace on the internet

Above is typical negotiation advice. If you ask for more than you want, the recruiter may meet you halfway.

It seems logical and great, but here's why you shouldn't follow that advice.

Haitian hostage rescue

In 1977, an official's aunt was kidnapped in Haiti. The kidnappers demanded $150,000 for the aunt's life. It seems reasonable until you realize why kidnappers want $150,000.

FBI detective and negotiator Chris Voss researched why they demanded so much.

“So they could party through the weekend”

When he realized their ransom was for partying, he offered $4,751 and a CD stereo. Criminals freed the aunt.

These thieves gave 31.57x their estimated amount and got a fraction. You shouldn't trust these thieves to negotiate your compensation.

What happened?

Negotiating your offer and Haiti

This narrative teaches you how to negotiate with a large number.

You can and will be talked down.

If a recruiter asks your wage expectation and you offer double, be ready to explain why.

If you can't justify your request, you may be offered less. The recruiter will notice and talk you down.

Reasonably,

• a tiny bit more than the present amount you earn

• a small premium over an alternative offer

• a little less than the role's allotted amount

Real-World Illustration

Recruiter: What’s your expected salary? Candidate: (I know the role is usually $100,000) $200,000 Recruiter: How much are you compensated in your current role? Candidate: $90,000 Recruiter: We’d be excited to offer you $95,000 for your experiences for the role.

So Why Do They Even Ask?

Recruiters ask for a number to negotiate a lower one. Asking yourself limits you.

You'll rarely get more than you asked for, and your request can be lowered.

The takeaway from all of this is to never give an expected compensation.

Tell them you haven't thought about it when you applied.