Models
Open vision models that plug into BOAT's editor. Compare what they do and what they need, then install one with a single command.
7 models · synced Sep 26, 2026
SAM3 LiteTextby EfficientSAM3
SAM3 with a distilled MobileCLIP text encoder. Keeps SAM3's text, box and point prompts while using less memory — the best default for most machines.
DetectionInstance seg.Semantic seg.Oriented boxesS0S1LSAM3by Meta
Meta's Segment Anything Model 3. Promptable concept segmentation from text, boxes or points — the most complete toolset in the library.
DetectionInstance seg.Semantic seg.Oriented boxesvit-hPET-DINO
Visual-prompt detector: draw one box and it finds every similar object. Pairs with Cascading for fast, repetitive detection work.
Detectionswin-tLocateAnythingby NVIDIA
NVIDIA's 3B grounding model, quantized to GGUF and run with llama.cpp on Apple silicon. Text to boxes, or text to points for counting crowds.
DetectionCounting3b-Q4_K_M3b-Q5_K_M3b-Q6_K3b-Q8_03b-BF16S3OD
Lightweight salient-object segmentation. Draw a loose box and it snaps to the object inside — as a mask or a rotated box — in under 1 GB of memory.
Instance seg.Semantic seg.Oriented boxesbasedissodLLMDetby iSEE Lab
Open-vocabulary detector trained with large-language-model supervision. Type what you are looking for and get bounding boxes.
Detectiontransformers-largemmdet-tinymmdet-tiny-p5mmdet-basemmdet-largeFalcon Perceptionby TII
TII's compact 600M-parameter perception model. Turns short text prompts into boxes and masks with a small memory footprint.
DetectionInstance seg.Semantic seg.600m
No models match those filters
Try fewer filters or a different search term.
Data from models.boat-annotation.com/index.json, snapshotted at build time.