Simple Vector HiDream
This HiDream LoRA is Lycoris based and trained to replicate vector art designs and styles, this LoRA leans more towards a modern and playful aesthetic rather than corporate style but it is capable of doing more than meets the eye, experiment with your prompts.
I recommend using LCM sampler with the simple scheduler, other samplers will work but not as sharp or coherent. The first image in the gallery will have an embedded workflow with a prompt example, try downloading the first image and dragging it into ComfyUI before complaining that it doesn't work. I don't have enough time to troubleshoot for everyone, sorry.
Trigger words: v3ct0r, cartoon vector art
Recommended Sampler: LCM
Recommended Scheduler: SIMPLE
Recommended Strength: 0.5-0.6
This model was trained to 2500 steps, 2 repeats with a learning rate of 4e-4 trained with Simple Tuner using the main branch. The dataset was around 148 synthetic images in total. All of the images used were 1:1 aspect ratio at 1024x1024 to fit into VRAM.
Training took around 3 hours using an RTX 4090 with 24GB VRAM, training times are on par with Flux LoRA training. Captioning was done using Joy Caption Batch with modified instructions and a token limit of 128 tokens (more than that gets truncated during training).
I trained the model with Full and ran inference in ComfyUI using the Dev model, it is said that this is the best strategy to get high quality outputs. Workflow is attached to the first image in the gallery, just drag and drop into ComfyUI.