ArtSynthLab

@AiHacksHQ Have found @Kling_ai to be more responsive to these more abstract images

Out take from my latest short. Animated with @Kling_ai #AIart #GenAI #aivideo

@AiHacksHQ Very cool! I always find the big video models have trouble bringing human animation to 'non' human like images. This is cool.

New Piece - Green Soft Porcelain Images - Midjourney V6 Motion - @Kling_ai Sound + Music - Ross Flight @bosskite Would love to know what you think it's about. #AIart #AIfilm #GenAI #kling #midjourney

@Koutchinski Fantastic how the code starts with 666!

Hello Darkness my old friend This style isn't inherently evil, but it's marked by intense contrast and deep, dark tones. Definitely one of my top 5—maybe even top 3—when it comes to graphic design aesthetics. Also, I hear ya. Done with Video. :-) --sref 66638475

@Kyrannio Still love a cheeky V5.2 when 6 ain't hitting right

I love this old Midjourney 5.1 vibe :). Feeling nostalgic today. From the archives

@burconsult 🙏🏻

Keep exploring 😍

@icreatelife Does less than 250 followers count? Would love a share

An 𝕏 algorithm has changed. Smaller accounts are boosted. Bigger accounts aren’t shown as much. It’s time for smaller accounts to shine!

@joehenriod @HBCoop_ @FAL @Magnific_AI @runwayml Hey thanks so much @joehenriod! You've inspired me to start posting a whole bunch vids that didn't make the cut, or might try and squeeze them into a new edit if I can find a narrative

@HBCoop_ @FAL @Magnific_AI @runwayml I've got two that were both so good they made me stop Tweeting and go for a walk. #1 x.com/ArtSynthLab/st…

✨Weekly Showcase✨ Share your favorite image, video, or project of the week. I'll go first: Images - FLUX Realism LoRA on @FAL Enhance / Upscale - @Magnific_AI Animation - @runwayml Gen-3 Music - Epidemic Sound Video Editing - CapCut

@joehenriod @midjourney @runwayml @bosskite Hey thanks so much!

@ArtSynthLab @midjourney @runwayml @bosskite This is the best thing EVER.

Bubblegum Baphomet Images - @midjourney V6 Motion - @runwayml Gen2 Sound - Ross Flight @bosskite

@joehenriod @occultbot

Do you believe Magick or The Occult can boost your performance in the "X" algorithm? Sacrifice an image to Baphomet in the comments below and find out.... Prompt Share: In an ornate room, a group clad in black robes gathers for an occult ritual. Baroque geometric patterns adorn the space, while large circular designs cover the floor. Flickering candles cast eerie shadows, creating a dramatic and intense atmosphere. The scene exudes an air of mystery and ancient mysticism. --ar 3:4 --sref 14544 --p --stylize 1000 --v 6.1

@nickfloats What we need is a style tuner but with an easy GUI of dials and sliders

If you're curious how the style reference system in Midjourney works, here are some notes from my conversation with their CEO on the topic: >What is Style Space? Midjourney’s style space is a multidimensional system containing a wide range of aesthetics. When you use --SREF random, the system selects a random point within this space. Each point represents a unique blend of styles--random hybrids of existing styles within the system, not tied to real-world artistic movements. >How it Works In Midjourney, a seed determines the starting randomness for generating an image. Similarly, –SREF random codes generate randomness for style within the style space, acting as coordinates that point to specific style locations. This space has hundreds of dimensions, offering endless possibilities for defining styles through colors, textures, lighting, and other features. >SREF System --sref codes and image references focus on adjusting style without changing the subject. This allows you to maintain the core elements of your image while modifying aspects like lighting or texture. The goal is to separate style from content for more precise control. >Personalization vs. SREF Personalization is broader and learns your personal preferences (e.g., colors, subjects), while SREF provides precise style control using specific references. >Community Response The community favors image based SREFs over codes because they offer a more visual and intuitive way to select styles, especially on the web app. Image based references make the process more interactive.

More experiments with @Kling_ai Loving the liquid aspects to this and keeping of the surreal quality of the original image. #AIart #aivideo

Started experimenting with @Kling_ai for video motion of my @midjourney images. Seems to have better motion for more abstract base images than runway. Keen to explore more. #AIart #aivideo #facemelt

Playing with my own video inputs to drive #animatediff #ComfyUI #depthanything

@BarkworthKnight Good discord for animatediff discord.gg/ksEjZYNH

@ArtSynthLab This is great. How you finding comfyUI? I've been using automatic1111 but about to swap over to comfyUI. Also have you found a way to make animatediff stay on the same image and not morph or evolve?

#ComfyUI #AnimateDiff

@BarkworthKnight Previously only used online services but moved to comfy for the modular experiemental approach. AnimateDiff seems pretty flexible, while not exactly filmic it can definitely be used for sublte movement

@nickfloats Only just discovered shift + click - drag to select - add to folder. Crazy useful.

Midjourney organization tip: when you're working in the alpha, don't forget to like your fav gens as you go click any image & hit like, or right click from the grid & like there then you can filter by "liked" in your archive makes finding the good stuff wayyy easier & faster