Trigger = DCVHSLMNLCR,
v1.0 Krea2Raw Trained @10,000 Steps, Rank 32, Captioned with Gpt4o, Dataset of 726 Images. It works perfectly with Krea2Turbo [GENERATE IMAGES AT 1mp "anything over 1mp resolution krea takes over and it loses the style] - Best to generate at 1mp [2k and 4k resolutions krea takes over too much] NEGATIVE: The Style did not stick well like moss did but thought i would release it anyway, I'm over training it done 3 different training runs and tested all steps and this version had the better output out of the lot, If i find a solution to better outputs ill post the v2.0 lora.
v1 MossBase Trained @7000 Steps, Rank 32, Captioned with Gpt4o, Dataset of 726 Images. It works perfectly with MossSFT.
DC Cursed VHS LiminalCore Helps to create the old style of @deadtempovisions on tiktok, The look of eerie VHS stills, with faded colors, soft focus, and deep shadows that make familiar places and strange figures with a dreamlike and unsettling Horror/Fantasy vibe.
Prompting Tips: It works better if you prompt like this [Trigger], [Subject + appearance + pose or action]. [Main visual element + how it lights the subject]. [Surroundings + framing + depth]. [Color palette + atmosphere]. [Overall lighting + contrast]. [Mood + what the image should make the viewer feel].
Other Keywords to use: eerie, surreal, dark, unsettling, dreamlike, otherworldly, mysterious, unease, horror, isolation, muted, ethereal.
Check the bottom of this lora page for more examples of this lora or check my profile gallery to see more examples and loras.
All images have prompts and comfyui workflows
If you use this lora in merges or upload to other generation sites please mention my civitai page on your post, Thankyou.
https://civitai.com/user/dkpc69