Currently only Flux.1 [dev], but other Flux models could be AnyFlow distilled into an n-step model.
Bagheera PRO
bghira
AI & ML interests
None yet
Recent Activity
posted an update 6 days ago
Generated ~44 million Cosmos Reasoning captions for CC12M, now available here: https://huggingface.co/datasets/webshart/cc12m-structured-captions
- webshart format, indexed with aspect bucketing metadata
- four captions per sample
- simple VLM output
- verbose caption
- a concise caption produced by passing the verbose caption through the model again to remove speculation and flowery language
- an ideogram4 style JSON schema
this used 8x H100 over about 6 days. published a dataset 6 days ago
webshart/cc12m-structured-captions updated a dataset 6 days ago
webshart/cc12m-structured-captionsOrganizations
Currently only Flux.1 [dev], but other Flux models could be AnyFlow distilled into an n-step model.
Demonstrating SimpleTuner's AnyFlow support by distilling Anima v1.0 using e621 dataset. Recommended weights for inference are dmd-v3 or dmd-v5.
v-prediction SDXL clone with zero-terminal SNR noise schedule
SD 2.x models with zero terminal SNR noise schedule.