文件清单 线路:原始(huggingface.co) · 切换
下载后放进对应的 ComfyUI 目录。models/ 指 ComfyUI 的模型根目录。
说明
Seedream style for H3
作者:nebsh
I really like Seedream 5.5 — it's currently my favorite closed-source image model, especially the way it balances contrast, lighting and grain. I was curious to see how much of that look I could bring to MiniMax H3 Image with a style LoRA, so I trained this experiment. Special thanks to @StefTranquillin for donating 47 images, and to @GrootDEX for providing around 100 images! Training happened in two stages: • First, 2,000 steps on the initial small dataset: 43 selected images. • Then I resumed that checkpoint and trained to 6,000 total steps using an expanded dataset of 1,134 images, including the original set. To use it with H3 Image: • Trigger: seedreamstyle. • LoRA strength: 1.0 for these tests. • Comparison settings: LCM, 6 steps, CFG 1. I've prepared a 20-scene gallery comparing H3 Without lora, the 2K checkpoint, the 6K checkpoint, and actual Seedream 5.5 generations. H3 examples were generated at 40 MP; Seedream references at native 2K. The few prompt-adjusted 6K retakes are clearly marked.
