improve: richer garment description + stronger prompt + 16 steps + 2MP ref- describe_garment now detects sleeve type, neckline, and special features (tie/bow, buttons, ruffles, etc.) using CLIP zero-shot — gives the model a much more specific caption to work from- Prompts now explicitly instruct the model to preserve sleeve shape, neckline, silhouette, and decorative elements without simplifying- Steps: 12 → 16 (better detail at ~25% more inference time)- Clothing reference: 1.5 → 2.0 megapixels (more detail in the conditioning)

feat: increase steps to 12 for better fabric texture

feat: increase inference steps to 8

feat: 150s GPU duration, increase inference steps 4→6

fix: add libgles2 system dep + fix SAM3 text prompt for clothing detection

initial