Prep captions for LoRA training: trigger word + content-only 47fb1ee goumsss Claude Opus 4.8 commited on Jun 12
Switch dataset model to Qwen-Image + add 54-image training set (LFS) 38bba68 goumsss Claude Opus 4.8 commited on Jun 12
Dataset: FLUX.2-dev + app-aligned prompts with multi-animal/place combos 18bda20 goumsss Claude Opus 4.8 commited on Jun 11
Upgrade image model to FLUX.2-klein-4B + align training pipeline 0a993a9 goumsss Claude Sonnet 4.6 commited on Jun 11
Add training dataset generator + fix MPS dtype for local inference 3311574 goumsss Claude Sonnet 4.6 commited on Jun 10
scripts: remove hardcoded paths, clean up run.sh kill logic d34e1e5 goumsss Claude Sonnet 4.6 commited on Jun 9
scripts/run.sh: kill all stale instances before starting 45af17f goumsss Claude Sonnet 4.6 commited on Jun 9
goumsss Claude Sonnet 4.6 commited on 
