LocateAnything delivery-vehicle detector

Drop an image to test the fine-tuned model. Results include native 0–1000, normalized 0–1, and original-image pixel coordinates.

Drop an image here

Detected objects

Classes to detect

Separate classes with commas, semicolons, or new lines.

Generation mode

Hybrid is recommended.

Maximum output tokens

128 2048

Drop an image to begin.

Detection results