Hugging Face
Models
Datasets
Spaces
Buckets
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
akhaliq
/
MOSS-Image-IDM-2-Steps
Like
12
Text-to-Image
English
Chinese
moss-image
lora
text-rendering
distillation
few-step
dmd
idm
ml-intern
arxiv:
4 papers
License:
apache-2.0
Model card
Files
Files and versions
xet
Community
Copy to bucket
main
MOSS-Image-IDM-2-Steps
/
training
89.7 kB
Ctrl+K
Ctrl+K
1 contributor
History:
11 commits
akhaliq
HF Staff
load_arm_lora: handle mixed-rank LoRAs via zero padding
73cf8d6
verified
about 1 hour ago
v2
load_arm_lora: handle mixed-rank LoRAs via zero padding
about 1 hour ago
edit_2step_eval.py
13.5 kB
Reference implementation: 2-step + armV5 editing at CFG 2 (ai-toolkit conditioning)
about 9 hours ago
rapid_moss_train_idm.py
25.4 kB
IDM method for rapid_moss (data-space): DMD direction + L_pop moment matching, no adversarial
about 13 hours ago
train_v3_idm_lora.py
11.9 kB
EMA swap via .data (fix autograd version-counter error in backward)
about 19 hours ago
