import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("OpenMOSS-Team/MOSS-Image-Base", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("lingjiu09/firefly_MOSS")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Firefly (Honkai: Star Rail) LoRA for MOSS-Image
📖 Model Description
This is a character LoRA for Firefly from Honkai: Star Rail, trained on the MOSS-Image-Base model. The dataset consists of 31 half-body and portrait images. It is optimized for generating facial details and upper-body compositions.
⚠️ Limitations
- No Full-Body Data: Since the current dataset only contains half-body images, generating
full bodyor wide shots may result in anatomy errors or out-of-character outfits. Please stick to upper-body prompts for the best results.
🙏 Credits
- Character: HoYoverse (miHoYo)
- Base Model: OpenMOSS-Team
- Training UI: Fizgig
🇨🇳 中文说明
📖 模型简介
这是一个基于 MOSS-Image-Base 训练的《崩坏：星穹铁道》角色——流萤 (Firefly) 的 LoRA 模型。 目前的数据集包含 31 张半身/特写图片，主要针对生成脸部细节和上半身构图进行了优化。
🎯 触发词
请务必在提示词开头加上专属触发词：
👉 firefly26710
⚠️ 注意事项
缺失全身数据：由于当前数据集仅使用半身图训练，强行生成全身图 (
full body) 可能会导致人体比例崩坏或服装错误。建议仅用于上半身生成。角色版权：HoYoverse (miHoYo)
底模来源：OpenMOSS-Team
训练工具：Fizgig
- Downloads last month
- 162
Model tree for lingjiu09/firefly_MOSS
Base model
OpenMOSS-Team/MOSS-Image-Base