VLM-AutoYOLO v1.5.3
What's Changed
- fix: fetch full detection detail on history select for mask display (a16f832)
- style: ruff format detection_strategy.py (264b2c9)
- fix: add missing modelType field in useYoloValidation Detection object (c73ff84)
- feat: batch UX, Docker SAM3, and core tests (188dcd5)
- style: increase sidebar width 420→440px, padding 1→1.25rem (38c26b8)
- docs: update structure and user guides for SAM3 integration (db5e421)
- docs: update HF_TOKEN guide - Read token inherits account permissions (2c455ea)
- docs: add HF_TOKEN requirement to SAM3 sections in README (EN/ZH) (632da47)
- chore: remove accidentally committed db and test images from git (fead419)
- fix: SAM3 unload fallback via port when process reference is lost (bca2f8e)
- feat: integrate SAM3 text-driven detection + segmentation strategy (7a7f2ac)
- docs: update Windows RTX 3080 10GB benchmark with clean-state results (53967ab)
- docs: update Mac model to MacBook Pro M4 Pro (281f36c)
- docs: add comprehensive macOS benchmark metrics (5f0f3c6)
- test: add detailed benchmark script with macOS memory tracking support (eb7e1a1)
Installation
Docker (Recommended)
curl -O https://raw.githubusercontent.com/Somnusochi/VLM-AutoYOLO/master/docker-compose.yml
docker compose up -dAccess:
- Frontend: http://localhost
- API Docs: http://localhost:8000/docs
Docker Images
docker pull ghcr.io/Somnusochi/VLM-AutoYOLO-frontend:v1.5.3
docker pull ghcr.io/Somnusochi/VLM-AutoYOLO-backend:v1.5.3Manual Setup
See README for detailed instructions.
Full Changelog: v1.5.2...v1.5.3