ViewTube

In case you didn't know Stable Diffusion 3 is now available to download, that is the 2B medium model.  https://huggingface.co/stabilityai/stable-diffusion-3-medium/tree/main  

Currently there is support in Comfy UI (workflows here)  https://huggingface.co/stabilityai/stable-diffusion-3-medium/tree/main/comfy_example_workflows  OR very basic  https://comfyanonymous.github.io/ComfyUI_examples/sd3/ 

You can also run it on Stable Swarm which is a front end with a backend that uses Comfy UI. I will do a video on this on the weekend.  https://github.com/Stability-AI/StableSwarmUI . Very easy to install.

With that being said, so far the feedback hasn't been great. I'm still testing things for myself and other than the "text" coherence, let's just say waiting for the fine tune models would be better.

Have you tried it? What are your first impressions? Let me know in the comments! 

Thank you MM fam! Even though it's taken me 4 years to get here I couldn't be happier! Not a lot of creators even get to 1K subs and getting passed 10K is only a 4% chance. Not to mention Playground's Channel has grown to 45K in such a short time as well.

I've shared before that my goal was to get to 10K and see what happens. Well here we are and I'm still loving it and not going anywhere for now.

Always appreciate the love and support and thank you....more than you know! 

Hey folks! I want to test the prompt adherence for SD3 in the next video, I'd appreciate any prompts you want me to try out! Or if there is anything else you want me to focus on let me know in the comments! 

Will you subscribe to Stability AI's Stable Assistant? More on it here  https://stability.ai/stable-assistant 
Some things to highlight...
-Access to SD3
-Chatbot Assistant, an advanced language model Stable LM2, for things like scripts, blogs and other text generation
-Image services, search and replace, remove background, control structure, sketch, creative upscale, outpaint
-Stable Video

Let me know your thoughts in the comments! 

Here it is officially (kind of)! SD3 via API at the moment, not locally just yet. Although if you read the post it states "While the model is available via API today, we are continuously working to improve the model in advance of its open release. In keeping with our commitment to open generative AI, we aim to make the model weights available for self-hosting with a Stability AI Membership soon".

You can sign up for a free membership to use locally when available but non-commercial. Otherwise a subscription is required to use commercially.

Thoughts? 
 https://stability.ai/news/stable-diffusion-3-api 

For those of you who have tried Udio and Suno AI which do you prefer? Let me know why in the comments! 

Proud of this, thank you all!

If I'm being honest, I admit, the last few weeks, I've been a bit demotivated. Don't get me wrong, I'm happy where I've been able to take this channel and surpass my own expectations but every now and then, doubt tends to creep in. Typically, I let it in, feel it, address it then move and I'm fine. But for whatever reason, these past few weeks, it's been a bit harder to shake off.

There is so much I want to do but can't, whether it's because I work full time, don't have the equipment I think would help me, or maybe I just want to be more creative with my content.....it can be a number of things any given day. It's definitely not burn out...maybe just a rut that we all go through, whatever it is....I have some things to think about.

That doesn't change the fact of how much I appreciate you all. Your comment, like, subscription, view and yes even the trolls, makes it all worth while.

To the next 2Mil and more, I'll see you when I see you. 

A bit different than what I normally post here but I wanted to share this from a fellow creator. We started roughly around the same time and came across her channel by chance. For those of you who happen to be part of my journey earlier on, you know that photography is a passion of mine and I hope to some day get back to it but more as a hobby I can just enjoy.

In any case If you are all about supporting smaller creators, and into photography/videography, maybe you might have something in common with Angela. I can tell you, she's real and genuine which is becoming very rare on the YT space these days. 

Perhaps it's me but I'm getting pretty annoyed how loosely the term "character consistency" has become. Midjourney just released their version and a couple months back Scenario also released their own version as well.

While yes both will give the character consistent faces and similar attire, consistent clothing is still flawed. You could pretty much do the same thing with Controlnet IP Adapter or reference. This isn't anything new. They may have their own backend pipeline but it still is not true consistency.

And of course every AI creator on YT is going to create content about "Midjourney Character consistency is here" or "Character consistency achieved with Midjourney" or however misleading way they want to lead you to click on the video to get your views...and views they will get.

At the end of the day, any progress towards character consistency is amazing! But I feel like there is a lot of misleading ideas and concepts around this topic. What are your thoughts? 

Happy to report Stable Cascade through Comfyui can run decently at 24 seconds on my 3060Ti. Basic set up but so far so good. Way better than the gradio version I tried a couple weeks back...5min! hahaha! Vid to come soon!