"The Name Of The Game Is Power" - AI Music Video Creation Process: 100% AI visuals, 100% human-made music Music: "Gone In 60 Seconds" by Mark DK Berry (available from https://markdkberry.bandcamp.com/trac... ) Introduction With the release of Wan 2.1 i2v (image-to-video) model for Comfyui in early March 2025, it opened up a whole new world of creative possibility. The track "Gone In 60 Seconds" seemed like a good choice for my next music video to test it out. Hardware & Setup **Equipment**: RTX 3060 (12GB VRAM), Windows 10, 32GB system RAM **Required Installations**: Sage attention, Triton, and Teacache **Note**: First sage attention install killed my Comfyui installation and required rebuilding (blessing in disguise as my setup had become bloated) Workflow Overview I spent a few days researching workflows then picked the best I could find. The workflow is available here: https://comfyworkflows.com/workflows/... I ended up using the same i2v workflow almost exclusively because it included interpolation and upscaling, which cut down post-production time. When making previous AI music videos, I limited myself to 5 days, but with image-to-video I aimed for better quality and more accurate depictions, giving it more time. I aimed at 10 days but managed to complete this version in 8. Image Creation Process 1. Used Flux-dev-fp8 in Comfyui to create base images (1344 x 768) 2. Used Krita with ACLY AI plugin (SDXL and Flux model) for segmenting, inpainting, and tweaking 3. Created my own Lora in Flux (3-hour process) to avoid copyright issues after previous experiences 4. Shotcut (or Topaz) for interpolating from 16 to 24 fps (I didnt use the Topaz enhancement features as I didnt feel it helped much after testing; it just improved the clarity of the underlying digital gremlins). Prompt Engineering Once input images were ready, I used AI assistants (Claude, Grok, or ChatGPT) to generate prompts based on criteria from a prompt-extender developed by Wan developers. This worked much better than simple 3-sentence requests (see the workflow link for that info https://comfyworkflows.com/workflows/... ). Video Generation Used Wan 2.1 i2v workflow to create 3-second video clips (16fps with upscale and interpolation) Most clips required 3-4 attempts, but some (especially car & motorcycle acceleration shots) took 20+ attempts Tried multiple Wan models but all struggled with moving vehicles while keeping other elements stationary Seed changes resolved most issues Post-Production Assembled clips in Davinci Resolve Spent time replacing unusable shots Acceptance & drawing a line in the sand Lessons Learned **Organization**: This process required tracking much more information than before, including shot-naming conventions and multiple CSV sheets to manage good/bad takes. When lipsync and ambient audio comes along, this is going to be challenging work for one person. **Flexibility**: The story I began with wasn't the story I ended up creating - it's best to start with a plan but allow AI to redirect it **Quality vs. Storytelling**: Sometimes sacrificing visual quality for storytelling works better, especially given hardware limitations Final Thoughts This AI model really feels like a leap ahead into a new dawn in visual storytelling. Just like in the 1920s when silent movies emerged, we are in the AI equivalent in 2025. Perhaps independant artists and activists will even stand a chance of challenging the Hollywood & Netflix narratives within a year or two. Follow and support me on my social media accounts as I continue this journey into AI music video creation. I hope you enjoyed this one. https://www.markdkberry.com https://markdkberry.bandcamp.com/ IG @ markdkberry X @ markdkberry #aimusicvideo #ai #music #musicvideo #markdkberry #wanx #comfyui

"I'm Still Alive (In The Naked Disco Of My Mind)" - Music Video This is the first finished track from an album of spoken word observations called "Modern Life" hopefully it will be fully released this year. The song is available from https://markdkberry.bandcamp.com/ This song is the first finished from the album "Modern Life". For the video, I used the old montage method I used to use before AI, of cutting movies into it. But then I tried to add myself in them, which was kind of fun. I used a n android phone and drape, and then green-screened myself using a free tier RunwayML login about a year ago. But I never liked it. It felt unfinished. So I recently used Comfyui and an anime video model to convert it all to anime and then used Davinci to overlay it onto the original in various opacities. It may get pulled, if so, then visit my Rumble channel https://rumble.com/user/mdkb and failing that Vimeo - https://vimeo.com/markdkberry Follow and support me on my social media accounts as I continue this journey into music video creation. I hope you enjoyed this one. markdkberry.com markdkberry.bandcamp.com IG @ markdkberry X @ markdkberry #aimusicvideo #ai #music #musicvideo #markdkberry #fearandloathinginlasvegas #eyeswideshut #oneflewoverthecuckoosnest

The music is 100% made by humans, the footage is 100% made using local AI. The soundtrack is available on https://markdkberry.bandcamp.com Equipment used for making AI footage: Hunyuan fp8 model with Fastvideo lora & tea cache on Comfyui running on a Windows 10 PC with 3060 RTX 12GB VRAM. Then it was all put together using Davinci. I ran out of time for color grading. I applied a new workflow on this project, and it gave me better quality and was faster. This was due to switching out the Hunyuan model and lowering the steps while keeping the Fastvideo lora in. I also used Tea cache. I decided to set it in a different aspect ratio more suitable to phone devices this time too. My workflow can be found here - https://comfyworkflows.com/workflows/c8d2c72d-76ea-40b1-9c0a-b6c064a88d95 It took about 4 days total but I don't have time to give it a color grading, which I normally would if I set a 5 day limit. The uncensored version of this video can be viewed here - https://sendvid.com/a7vibclk Follow and support me on my social media accounts as I continue this journey into AI music video creation. I hope you enjoyed this one. markdkberry.com markdkberry.bandcamp.com IG @ markdkberry X @ markdkberry #aimusicvideo #ai #music #musicvideo #markdkberry #shibari #garage #edm

"Baes With Guns" by Mark DK Berry (EDM Music video) BPM 133 The music is 100% made by human, the footage is 100% made using AI. The soundtrack is available on https://markdkberry.bandcamp.com Equipment used for making AI footage: Hunyuan GGUF model on Comfyui running on a Windows 10 PC with 3060 RTX 12GB VRAM. Then it was all put together using Shotcut and Davinci. I am still working on achieving better renditions at scale within 3 to 4 day time frames. It took about 4 days including writing the song, (5 days is where I set my limit on how long to spend on these currently). Most of that time is redoing stuff that doesnt work. A longer DJ version of this track will be available via my website at a later date. Follow and support me on my social media accounts as I continue this journey into AI music video creation. I hope you enjoyed this one. markdkberry.com markdkberry.bandcamp.com IG @ markdkberry X @ markdkberry #aimusicvideo #ai #music #musicvideo #markdkberry #baeswithguns #war weapons #guns

The music video is an homage to the original "The Planet of The Apes" with a mashup taken from the 1968 movie. "The Ape Never Will" (Melodic Tech House) - Mark DK Berry This track and others available for download and stream via - https://markdkberry.com

"Counterfeit Love" by Mark DK Berry (Melodic Tech House) This track and others available for download and stream via - https://markdkberry.com

"Fallen Angel" - A melodic house track and video by Mark DK Berry. The experimental video took three months to complete as a total novice learning Unreal Engine and was an attempt at animating Metahumans. The vocals were created using Mai - an AI vocal provided with Synth V pro. She is a little too Asian, but it worked with the track. Two biggest issues were that animation is not easy, and clothing Metahumans is expensive and hard to make it look right. So, I removed their clothes instead, and gave them mannequin features, which may or may not get banned by YT as a result. The animation was eventually done using Cascadeur rather than inside of UE 5.2 which I found too cumbersome. It was then imported back into UE, but I found it awkward to tweak and fix. I look forward to someone coming up with "text to animation" AI in this area, it needs it. I am not an iPhone user so I had no easy way to animate facial movement or lip sync, so I didn't bother. I admittedly plagiarized the video idea - if you can you guess the video and song I took the idea from mention in the comments. But doing this gave me the opportunity to focus on how to light things properly and camera movements. All of which I copied rather than worry about figuring out that part of it, which worked in some ways and didn't in others. But I was trying to learn so many things from scratch inside Unreal Engine involving lighting/animation/cameras etc... hence why it was 3 months instead of 3 weeks to complete. I am not exactly happy with the result and find it pretty cheesy, but it was fun doing it. I won't attempt another until the MoCap animation is better (android or webcam), and until animation can be done by AI from text prompts. But all the tools look promising, have real potential for making some amazing stuff, and I hope to use them again sometime in the future. The track will be up on release on major platforms for streaming soon. In the meantime it is available via my bandcamp here - https://markdkberry.bandcamp.com/track/fallen-angel-melodic-house along with the lyrics if you are interested. If you are a music video creator interested in AI and animation feel free to comment below, or ask any questions. I am interested in learning from anyone else who has ventured into this area and can share their methods. It wont be long before it is possible to make higher quality music videos this way. I had hoped to animate some of my album ideas, but it just isn't there yet. Equipment used: Windows 10 x64 with 32GB RAM (started with 16GB but increased it mid way through) RTX 2060 with 6GB VRAM (this was the biggest issue, you need more VRAM) Unreal Engine 5.2 (free) Cascadeur (free version) Davinci Resolve 18.5 (free) and yes, with that equipment it was barely able to complete the video, and I had a lot of crashes, but I was able to finish it. I was restricted to exporting at 1280x720 and 30-fps on that entry level gear.