obss /sahi
Framework agnostic sliced/tiled inference + interactive ui + error analysis plots
★ 5,604⑂ 793👁 47⑃ 891● 0▤ 120
+1
日增长
5,604
stars
793
forks
47
watchers
891
prs
0
issues
120
releases
README
SAHI: Slicing Aided Hyper Inference
A lightweight vision library for performing large scale object detection & instance segmentation
Overview
SAHI helps developers overcome real-world challenges in object detection by
enabling sliced inference for detecting small objects in large images. It
supports various popular detection models and provides easy-to-use APIs.
🌐 English | 🇨🇳 简体中文 | 🇹🇷 Türkçe
| Command | Description |
| ----------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| predict | Perform sliced/standard video/image prediction using any ultralytics / mmdet / huggingface / torchvision model, see CLI guide |
| predict-fiftyone | Perform sliced/standard prediction using any supported model and explore results in fiftyone app, learn more |
| coco slice | Automatically slice COCO annotation and image files, see slicing utilities |
| coco fiftyone | Explore multiple prediction results on your COCO dataset with fiftyone ui ordered by number of misdetections |
| coco evaluate | Evaluate classwise COCO AP and AR for given predictions and ground truth, check COCO utilities Discussions
#1508YOLOVAL: a two-image sliced-inference check with game-specific ONNX models (paid)0 comments · 2026-11-04#1476Responsible disclosure: requesting a private channel to report a security issue2 comments · 2026-10-15#1420🚀 SAHI 0.12.5: postprocessing no longer scales with how crowded your boxes are1 comments · 2026-08-04#1415Can we improve the memory usage of the slicing operation?2 comments · 2026-08-02#1413Proposal: Sparse CSR fast path for large Numba NMS and GreedyNMM workloads0 comments · 2026-08-01#1397Add support for Libreyolo2 comments · 2026-07-22#1396SAHI 0.12.2 is out: postprocessing no longer allocates a quadratic matrix0 comments · 2026-07-20#1388Inference for RF-DETRSeg Model1 comments · 2026-07-08
Latest release
0.14.0 0.14.0 2026-11-01A minor release that moves the tests, docs and examples to YOLO26, defaults YOLOE to YOLOE-26, fixes bugs in the YOLOE, YOLO-World, torchvision, Hugging Face and mmdet wrappers, downloads and scripts, and makes COCO slicing and mask drawing faster with the same output. 💥 Breaking Changes - YOLOE loads yoloe-26s-seg.pt when no model_path is given (1499) instead of yoloe-11s-seg.pt. Pass model_path to keep using YOLOE-11. - The hugging_face_universal_segmentation model type is removed (1498). It had no module behind it and always failed with ModuleNotFoundError. Use huggingface_segmentation f