SAVRN
Search Contact SAVRN

Open-weight model · Image segmentation

sae-2026-hang-all-yolo26n-seg-v2-960

by Black Bee Drones blackbeedrones/sae-2026-hang-all-yolo26n-seg-v2-960

Instance-segmentation model for the "hang the wire" hook mission of the SAE Brasil Eletroquad 2026 competition, trained by Black Bee Drones.

Parameters—
Context—
Weights3.8 GB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads3k

Model Card

By Black Bee Drones, published under apache-2.0, revision b0e189a9e9e5.

Instance-segmentation model for the "hang the wire" hook mission of the SAE Brasil Eletroquad 2026 competition, trained by Black Bee Drones. It segments two classes - rose (each visible segment of the suspended rope) and sphere (the orange sphere on the rope) - and runs on every control-loop tick of the mission. The team finished 2nd overall The drone takes off, finds the orange sphere mounted on one of two suspended ropes, parks a fixed distance from it, picks which side of the rope to fly along, turns perpendicular to the rope, descends on LIDAR, releases a hook with a servo, and lands. Because each visible rope segment is its own rose instance, the controller can measure both rope arms…

Read Black Bee Drones's full model card

SAE Eletroquad 2026 - Hook segmentation (YOLO26n-seg)

Instance-segmentation model for the "hang the wire" hook mission of the SAE Brasil Eletroquad 2026 competition, trained by Black Bee Drones. It segments two classes - rose (each visible segment of the suspended rope) and sphere (the orange sphere on the rope) - and runs on every control-loop tick of the mission. The team finished 2nd overall (official results).

  • Base model: yolo26n-seg.pt (Ultralytics)
  • Task: instance segmentation
  • Classes: rose (0), sphere (1)
  • Input size: 960
  • Dataset: blackbeedrones/sae-2026-hook

Mission

The drone takes off, finds the orange sphere mounted on one of two suspended ropes, parks a fixed distance from it, picks which side of the rope to fly along, turns perpendicular to the rope, descends on LIDAR, releases a hook with a servo, and lands. Because each visible rope segment is its own rose instance, the controller can measure both rope arms and choose a side. The mission runs on a Jetson Orin Nano through Nectar SDK.

Results

Evaluated on the dataset's test split (395 images, 501 instances), NMS iou=0.6. The numbers below are Ultralytics-native (COCO 101-point interpolation); see SUMMARY.md for the full breakdown, including the SDK/torchmetrics figures (which run a different curve discretization and read systematically lower at mAP@50-95).

Metric Box Mask
mAP@50 0.9929 0.9844
mAP@50-95 0.9640 0.8535

Per-class mAP@50-95:

Class Box Mask
rose 0.9450 0.7555
sphere 0.9831 0.9515

Operating-point precision / recall / F1 at the per-class optimal confidences (from the F1 curve):

Class conf Box P / R / F1 Mask P / R / F1
rose 0.47 0.963 / 0.975 / 0.969 0.946 / 0.958 / 0.952
sphere 0.70 0.987 / 0.993 / 0.990 0.987 / 0.993 / 0.990

Recommended inference settings

Setting Value
imgsz 960
iou (NMS) 0.6
conf (rose) 0.47
conf (sphere) 0.70

Predict at the lower of the two confidences and apply a per-class filter, so each class keeps its own threshold.

Usage

Nectar SDK

from nectar.ai.segmentation import Segmentor
from nectar.ai.detection.postprocess import PerClassConfidenceFilter

# loads weights/best.pt from the Hub
segmentor = Segmentor("blackbeedrones/sae-2026-hang-all-yolo26n-seg-v2-960")
segmentor.load()

result = segmentor.segment(image, conf=0.47, iou=0.6, imgsz=960)

# rose=0, sphere=1
per_class = PerClassConfidenceFilter(threshold_mapping={0: 0.47, 1: 0.70}, default_threshold=0.47)
kept = per_class.filter(result.to_supervision())
for seg in result:
    print(seg.class_name, f"{seg.confidence:.2f}", f"area={seg.mask_area}px")

Ultralytics

from huggingface_hub import hf_hub_download
from ultralytics import YOLO

weights = hf_hub_download("blackbeedrones/sae-2026-hang-all-yolo26n-seg-v2-960", "weights/best.pt")
model = YOLO(weights)
results = model.predict("image.jpg", imgsz=960, iou=0.6, conf=0.47)

Training

Trained with Ultralytics through the Nectar SDK on blackbeedrones/sae-2026-hook (Roboflow sae-2026-hang v2: 9,235 train / 396 val / 395 test images). Full configuration in experiment.config.yaml.

Parameter Value
Base model yolo26n-seg.pt
Image size 960
Batch size 48
Epochs up to 250, early-stopped at 165 (patience 20)
LR / schedule 0.01, linear with cosine (cos_lr), lrf=0.01
Weight decay 5e-4
Warmup 3 epochs
EMA on (decay 0.9997)
Augmentation mosaic 1.0 (closed last 10 epochs), HSV, fliplr 0.5, scale 0.5, translate 0.1
Mask overlap_mask=true, mask_ratio=4
Seed 42

Files

  • weights/best.pt - recommended checkpoint (best validation).
  • weights/last.pt, weights/epoch*.pt - last and per-epoch checkpoints.
  • evaluation/ - curves, confusion matrix, prediction samples, per-class CSV/JSON.
  • experiment.config.yaml, args.yaml, results.csv - training configuration and log.
  • SUMMARY.md - full evaluation breakdown and threshold analysis.

References

Identity and Version

Repository
blackbeedrones/sae-2026-hang-all-yolo26n-seg-v2-960
Publisher
Black Bee Drones
Task
Image segmentation
Modality
Image
Library
ultralytics
Parameters
Not stated by the source
Languages
uav
Revision
b0e189a9e9e57b1416e82cc00e2bb4d342062683
First published
2026-05-05
Last updated
2026-06-20

Files and Weights

316 files, 3.9 GB in total. The weights are 202 files totalling 3.8 GB in pt.

Weights202 files · 3.8 GB
Configuration20 files · 1.4 MB
Documentation2 files · 9.3 KB
Other91 files · 52.3 MB
Repository1 file · 4.4 KB
Every file
FileTypeSizeSHA-256
weights/best.ptWeights19.0 MB 8ce24b50f239
weights/epoch0.ptWeights19.0 MB b89803fd228d
weights/epoch1.ptWeights19.0 MB 2a00fbd9bfb6
weights/epoch10.ptWeights19.0 MB 30b2afeeaab8
weights/epoch100.ptWeights19.0 MB 2c924f947cdf
weights/epoch101.ptWeights19.0 MB f8eab4e6a52d
weights/epoch102.ptWeights19.0 MB e511f244017d
weights/epoch103.ptWeights19.0 MB 1d42c9576f20
weights/epoch104.ptWeights19.0 MB 122aa3c99fc6
weights/epoch105.ptWeights19.0 MB b0ed7d296862
weights/epoch106.ptWeights19.0 MB 9a96c877ca0f
weights/epoch107.ptWeights19.0 MB 0c4e75cef416
weights/epoch108.ptWeights19.0 MB f749bfd2e13e
weights/epoch109.ptWeights19.0 MB 0c482b95e87b
weights/epoch11.ptWeights19.0 MB f039b9df0870
weights/epoch110.ptWeights19.0 MB 1847ae645891
weights/epoch111.ptWeights19.0 MB 45d11ff5fce4
weights/epoch112.ptWeights19.0 MB bb518f7c79f0
weights/epoch113.ptWeights19.0 MB ed35f0eeb7ec
weights/epoch114.ptWeights19.0 MB 44d4478a274a
weights/epoch115.ptWeights19.0 MB 80dd668a872a
weights/epoch116.ptWeights19.0 MB b06b0456a43f
weights/epoch117.ptWeights19.0 MB 74cc432827d6
weights/epoch118.ptWeights19.0 MB 8114f4dc1ade
weights/epoch119.ptWeights19.0 MB a22273b201b8
weights/epoch12.ptWeights19.0 MB 96ace7e1c323
weights/epoch120.ptWeights19.0 MB ad6949ba6b37
weights/epoch121.ptWeights19.0 MB 2cb57c22e13f
weights/epoch122.ptWeights19.0 MB 629169045550
weights/epoch123.ptWeights19.0 MB bad389f4b9fc
weights/epoch124.ptWeights19.0 MB ec788b45d826
weights/epoch125.ptWeights19.0 MB 4d5f37969aca
weights/epoch126.ptWeights19.0 MB 4b043769639b
weights/epoch127.ptWeights19.0 MB 0025280c937c
weights/epoch128.ptWeights19.0 MB 3275d0b33c78
weights/epoch129.ptWeights19.0 MB 58c391006bf2
weights/epoch13.ptWeights19.0 MB 63bd13558b6d
weights/epoch130.ptWeights19.0 MB 9e15bbdb9ad3
weights/epoch131.ptWeights19.0 MB c8c225974d22
weights/epoch132.ptWeights19.0 MB 8401cfa2d1b0
weights/epoch133.ptWeights19.0 MB 7b87fc74363f
weights/epoch134.ptWeights19.0 MB 37f249dd1cc8
weights/epoch135.ptWeights19.0 MB d81a5f71f3ed
weights/epoch136.ptWeights19.0 MB 0634ded1646f
weights/epoch137.ptWeights19.0 MB f5b4f2fac53c
weights/epoch138.ptWeights19.0 MB 57743ca1b017
weights/epoch139.ptWeights19.0 MB 98358ec03afd
weights/epoch14.ptWeights19.0 MB 4ad23374a446
weights/epoch140.ptWeights19.0 MB b75479e9d0a4
weights/epoch141.ptWeights19.0 MB 0cb634eebf0b
weights/epoch142.ptWeights19.0 MB 2afe6afe1de2
weights/epoch143.ptWeights19.0 MB 4631d9411c3a
weights/epoch144.ptWeights19.0 MB 153d315df087
weights/epoch145.ptWeights19.0 MB 5edb21afce09
weights/epoch146.ptWeights19.0 MB b1314609b230
weights/epoch147.ptWeights19.0 MB fac963934dc8
weights/epoch148.ptWeights19.0 MB 1c040887e986
weights/epoch149.ptWeights19.0 MB d7d85467db1c
weights/epoch15.ptWeights19.0 MB 9771d2c9e590
weights/epoch150.ptWeights19.0 MB 7c3ac9f3482c
weights/epoch151.ptWeights19.0 MB acb3cc834f69
weights/epoch152.ptWeights19.0 MB f77a55033dc4
weights/epoch153.ptWeights19.0 MB 57ef0935436d
weights/epoch154.ptWeights19.0 MB 31b821779d33
weights/epoch155.ptWeights19.0 MB c666cf6dca84
weights/epoch156.ptWeights19.0 MB 28345a6878c1
weights/epoch157.ptWeights19.0 MB 834c474d03ba
weights/epoch158.ptWeights19.0 MB 4dc7c169b8e5
weights/epoch159.ptWeights19.0 MB a1bb425412e1
weights/epoch16.ptWeights19.0 MB a259151fe77b
weights/epoch160.ptWeights19.0 MB 3f8c5087b3ee
weights/epoch161.ptWeights19.0 MB b3791c2639be
weights/epoch162.ptWeights19.0 MB 8299a2c7e0d4
weights/epoch163.ptWeights19.0 MB eb5919ed4bb5
weights/epoch164.ptWeights19.0 MB 9df655a66052
weights/epoch165.ptWeights19.0 MB e6eabd07cb58
weights/epoch166.ptWeights19.0 MB f9d34c7ed58d
weights/epoch167.ptWeights19.0 MB 7b19ac65d6cb
weights/epoch168.ptWeights19.0 MB 956e9ded7c1b
weights/epoch169.ptWeights19.0 MB 5fd249598cde
weights/epoch17.ptWeights19.0 MB 77fddd6b82fa
weights/epoch170.ptWeights19.0 MB e0551a5df1b0
weights/epoch171.ptWeights19.0 MB 9f6cd465c993
weights/epoch172.ptWeights19.0 MB 29954b65e134
weights/epoch173.ptWeights19.0 MB 7aa799aecc50
weights/epoch174.ptWeights19.0 MB e6110551793c
weights/epoch175.ptWeights19.0 MB 217517836e56
weights/epoch176.ptWeights19.0 MB ee81dc185313
weights/epoch177.ptWeights19.0 MB 41f37890670f
weights/epoch178.ptWeights19.0 MB 63a498de1497
weights/epoch179.ptWeights19.0 MB 83fa5bec5744
weights/epoch18.ptWeights19.0 MB ab774b499870
weights/epoch180.ptWeights19.0 MB d1f66472970a
weights/epoch181.ptWeights19.0 MB 89826827a76b
weights/epoch182.ptWeights19.0 MB 87dc186fc89b
weights/epoch183.ptWeights19.0 MB 9b5ccf667150
weights/epoch184.ptWeights19.0 MB 494d992b5603
weights/epoch185.ptWeights19.0 MB 344fe95a2687
weights/epoch186.ptWeights19.0 MB 6932cab16e6b
weights/epoch187.ptWeights19.0 MB 6f82dfe0eb00
weights/epoch188.ptWeights19.0 MB f62621c5e38c
weights/epoch189.ptWeights19.0 MB c0eaf5ea66a4
weights/epoch19.ptWeights19.0 MB 3365ab6b5ae6
weights/epoch190.ptWeights19.0 MB fe3025d19de2
weights/epoch191.ptWeights19.0 MB 9c322c587c48
weights/epoch192.ptWeights19.0 MB 1416ff80ca13
weights/epoch193.ptWeights19.0 MB 15c25635f1d2
weights/epoch194.ptWeights19.0 MB a1fcb9d9c3df
weights/epoch195.ptWeights19.0 MB a7d59de86df2
weights/epoch196.ptWeights19.0 MB e782bb55ac35
weights/epoch197.ptWeights19.0 MB 247c29166878
weights/epoch198.ptWeights19.0 MB c6f87707ce3a
weights/epoch199.ptWeights19.0 MB 8ce24b50f239
weights/epoch2.ptWeights19.0 MB 36e0255cd377
weights/epoch20.ptWeights19.0 MB 5b4fd47576bf
weights/epoch21.ptWeights19.0 MB 49d9b5a7ec34
weights/epoch22.ptWeights19.0 MB 6492ac1b0dfa
weights/epoch23.ptWeights19.0 MB 3a2492403b02
weights/epoch24.ptWeights19.0 MB 1a794b83e5f0
weights/epoch25.ptWeights19.0 MB aa12b607f595
weights/epoch26.ptWeights19.0 MB 27d9d2429c85
weights/epoch27.ptWeights19.0 MB 1a3056458795
weights/epoch28.ptWeights19.0 MB dc95921b89b5
weights/epoch29.ptWeights19.0 MB 9c85dd6aba2a
weights/epoch3.ptWeights19.0 MB 3c9ab6ba8c24
weights/epoch30.ptWeights19.0 MB daddf6e5efd9
weights/epoch31.ptWeights19.0 MB 6f480d32e62b
weights/epoch32.ptWeights19.0 MB 82d6bca1d2f3
weights/epoch33.ptWeights19.0 MB 9e0b282b149a
weights/epoch34.ptWeights19.0 MB 9e61611242d2
weights/epoch35.ptWeights19.0 MB 253d5e6f4e88
weights/epoch36.ptWeights19.0 MB f4f4a018f127
weights/epoch37.ptWeights19.0 MB 4105640f32ef
weights/epoch38.ptWeights19.0 MB c5bf2efa5482
weights/epoch39.ptWeights19.0 MB 85d5bc2e4472
weights/epoch4.ptWeights19.0 MB e8566606a275
weights/epoch40.ptWeights19.0 MB 5c2780cc1373
weights/epoch41.ptWeights19.0 MB d11d10ea922a
weights/epoch42.ptWeights19.0 MB 9980989e9aa0
weights/epoch43.ptWeights19.0 MB ff5c2f873b94
weights/epoch44.ptWeights19.0 MB c3d8748d9f3a
weights/epoch45.ptWeights19.0 MB e1f1d8e7b07d
weights/epoch46.ptWeights19.0 MB cab4ae62900f
weights/epoch47.ptWeights19.0 MB 6bde170849bd
weights/epoch48.ptWeights19.0 MB efc2fffb2abc
weights/epoch49.ptWeights19.0 MB 38e416861d4d
weights/epoch5.ptWeights19.0 MB 9ea11a8e060b
weights/epoch50.ptWeights19.0 MB a64ab84f3f9d
weights/epoch51.ptWeights19.0 MB 65db10886410
weights/epoch52.ptWeights19.0 MB fafee9ec4c58
weights/epoch53.ptWeights19.0 MB 58843fa3eee4
weights/epoch54.ptWeights19.0 MB f7cf3a075ba1
weights/epoch55.ptWeights19.0 MB 1758f66cd3cf
weights/epoch56.ptWeights19.0 MB ae7220144b49
weights/epoch57.ptWeights19.0 MB 25a251245dd1
weights/epoch58.ptWeights19.0 MB 58f594b51d00
weights/epoch59.ptWeights19.0 MB 6ce50dcf7cd9
weights/epoch6.ptWeights19.0 MB cf8b611d91ff
weights/epoch60.ptWeights19.0 MB 3303fc6f104d
weights/epoch61.ptWeights19.0 MB 924f7b7e7f9d
weights/epoch62.ptWeights19.0 MB de5c64b53fdd
weights/epoch63.ptWeights19.0 MB 7e2329083539
weights/epoch64.ptWeights19.0 MB 07f2025d92d2
weights/epoch65.ptWeights19.0 MB 1ab3d4fc6b9f
weights/epoch66.ptWeights19.0 MB 5a0a170fbc84
weights/epoch67.ptWeights19.0 MB 3b7356a2d110
weights/epoch68.ptWeights19.0 MB 1765b5aea103
weights/epoch69.ptWeights19.0 MB 248a3e4e54e0
weights/epoch7.ptWeights19.0 MB bffd4a757b5c
weights/epoch70.ptWeights19.0 MB 948afbada2f0
weights/epoch71.ptWeights19.0 MB 9640f73946ae
weights/epoch72.ptWeights19.0 MB 80c5d22014f2
weights/epoch73.ptWeights19.0 MB 13231c6e8416
weights/epoch74.ptWeights19.0 MB dcaa63b22b7b
weights/epoch75.ptWeights19.0 MB 43dc0391f7ef
weights/epoch76.ptWeights19.0 MB 3e031c6b6231
weights/epoch77.ptWeights19.0 MB 6614d9a2c4fe
weights/epoch78.ptWeights19.0 MB c40cb0686aba
weights/epoch79.ptWeights19.0 MB 855358b056d3
weights/epoch8.ptWeights19.0 MB 93d8b1354b67
weights/epoch80.ptWeights19.0 MB 5d1f2e0e72f7
weights/epoch81.ptWeights19.0 MB c26fa03a25a1
weights/epoch82.ptWeights19.0 MB 2729eec86cb5
weights/epoch83.ptWeights19.0 MB e4da54362018
weights/epoch84.ptWeights19.0 MB 8b23a71ad680
weights/epoch85.ptWeights19.0 MB 29d5d92053d7
weights/epoch86.ptWeights19.0 MB 9d4c8ee5168e
weights/epoch87.ptWeights19.0 MB 843c0e4a9d28
weights/epoch88.ptWeights19.0 MB 1479c187df9f
weights/epoch89.ptWeights19.0 MB b5fc7f029f23
weights/epoch9.ptWeights19.0 MB 43087abd575b
weights/epoch90.ptWeights19.0 MB 34f88fb9b2c8
weights/epoch91.ptWeights19.0 MB d8c8d7d513cf
weights/epoch92.ptWeights19.0 MB 5f0293fc29a6
weights/epoch93.ptWeights19.0 MB 60bf6b8c4843
weights/epoch94.ptWeights19.0 MB 724818488e08
weights/epoch95.ptWeights19.0 MB 52429b6b5692
weights/epoch96.ptWeights19.0 MB 346ac467eb4b
weights/epoch97.ptWeights19.0 MB 86fcc2adfabe
weights/epoch98.ptWeights19.0 MB 35980da7e584
weights/epoch99.ptWeights19.0 MB d0e91a44d522
weights/last.ptWeights19.0 MB e61ea15fb2fb
args.yamlConfiguration1.6 KB —
eval_sdk_default/evaluation_report.jsonConfiguration3.9 KB —
eval_sdk_default/evaluation_results.jsonConfiguration389.9 KB —
eval_sdk_default/metrics_summary.jsonConfiguration480 B —
eval_sdk_default/per_class_metrics.jsonConfiguration447 B —
eval_sdk_old_test/evaluation_report.jsonConfiguration3.9 KB —
eval_sdk_old_test/evaluation_results.jsonConfiguration209.8 KB —
eval_sdk_old_test/metrics_summary.jsonConfiguration479 B —
eval_sdk_old_test/per_class_metrics.jsonConfiguration439 B —
eval_sdk_optimal/evaluation_report.jsonConfiguration3.9 KB —
eval_sdk_optimal/evaluation_results.jsonConfiguration389.9 KB —
eval_sdk_optimal/metrics_summary.jsonConfiguration482 B —
eval_sdk_optimal/per_class_metrics.jsonConfiguration458 B —
evaluation/evaluation_report.jsonConfiguration3.7 KB —
evaluation/evaluation_results.jsonConfiguration389.9 KB —
evaluation/metrics_summary.jsonConfiguration483 B —
evaluation/per_class_metrics.jsonConfiguration442 B —
experiment.config.yamlConfiguration1.7 KB —
optimal_thresholds.jsonConfiguration1.5 KB —
preprocessing_ablation.jsonConfiguration2.5 KB —
README.mdDocumentation6.1 KB —
SUMMARY.mdDocumentation3.2 KB —
eval_sdk_default/BoxF1_curve.pngOther125.6 KB 5e12129b00d2
eval_sdk_default/BoxPR_curve.pngOther102.2 KB df46df138aed
eval_sdk_default/BoxP_curve.pngOther104.4 KB 4fc51da2f253
eval_sdk_default/BoxR_curve.pngOther125.9 KB 5a510342e8fc
eval_sdk_default/MaskF1_curve.pngOther125.5 KB 06d999c3138d
eval_sdk_default/MaskPR_curve.pngOther111.9 KB 12f9a64180e5
eval_sdk_default/MaskP_curve.pngOther110.9 KB 036d732b9a80
eval_sdk_default/MaskR_curve.pngOther125.1 KB 58fc1fc4d4cd
eval_sdk_default/confusion_matrix.pngOther40.5 KB —
eval_sdk_default/error_analysis.pngOther193.4 KB 865b9f6036b2
eval_sdk_default/error_statistics.csvOther246 B —
eval_sdk_default/evaluation_metrics.csvOther392 B —
eval_sdk_default/f1_metrics.csvOther287 B —
eval_sdk_default/mar_metrics.csvOther471 B —
eval_sdk_default/per_class_metrics.csvOther254 B —
eval_sdk_default/performance_analysis.pngOther89.6 KB —
eval_sdk_default/pr_analysis_results.csvOther304 B —
eval_sdk_default/precision_metrics.csvOther276 B —
eval_sdk_default/prediction_samples.pngOther13.6 MB 02d4a0cedac1
eval_sdk_default/recall_metrics.csvOther277 B —
eval_sdk_default/results.pngOther43.7 KB —
eval_sdk_old_test/BoxF1_curve.pngOther121.3 KB f485bd793c30
eval_sdk_old_test/BoxPR_curve.pngOther99.4 KB —
eval_sdk_old_test/BoxP_curve.pngOther104.1 KB 9f8e5e949f71
eval_sdk_old_test/BoxR_curve.pngOther124.0 KB bd5833682027
eval_sdk_old_test/MaskF1_curve.pngOther119.9 KB 935af14b2948
eval_sdk_old_test/MaskPR_curve.pngOther106.0 KB b31154843b3a
eval_sdk_old_test/MaskP_curve.pngOther108.0 KB f040c3538217
eval_sdk_old_test/MaskR_curve.pngOther119.5 KB e28dd428c163
eval_sdk_old_test/confusion_matrix.pngOther39.8 KB —
eval_sdk_old_test/error_analysis.pngOther188.1 KB 33ff70eda9c1
eval_sdk_old_test/error_statistics.csvOther214 B —
eval_sdk_old_test/evaluation_metrics.csvOther391 B —
eval_sdk_old_test/f1_metrics.csvOther287 B —
eval_sdk_old_test/mar_metrics.csvOther473 B —
eval_sdk_old_test/per_class_metrics.csvOther246 B —
eval_sdk_old_test/performance_analysis.pngOther90.0 KB —
eval_sdk_old_test/pr_analysis_results.csvOther303 B —
eval_sdk_old_test/precision_metrics.csvOther267 B —
eval_sdk_old_test/prediction_samples.pngOther15.8 MB 75d4257bd455
eval_sdk_old_test/recall_metrics.csvOther263 B —
eval_sdk_old_test/results.pngOther42.5 KB —
eval_sdk_optimal/BoxF1_curve.pngOther125.6 KB 5e12129b00d2
eval_sdk_optimal/BoxPR_curve.pngOther102.2 KB df46df138aed
eval_sdk_optimal/BoxP_curve.pngOther104.4 KB 4fc51da2f253
eval_sdk_optimal/BoxR_curve.pngOther125.9 KB 5a510342e8fc
eval_sdk_optimal/MaskF1_curve.pngOther125.5 KB 06d999c3138d
eval_sdk_optimal/MaskPR_curve.pngOther111.9 KB 12f9a64180e5
eval_sdk_optimal/MaskP_curve.pngOther110.9 KB 036d732b9a80
eval_sdk_optimal/MaskR_curve.pngOther125.1 KB 58fc1fc4d4cd
eval_sdk_optimal/confusion_matrix.pngOther40.5 KB —
eval_sdk_optimal/error_analysis.pngOther193.4 KB 865b9f6036b2
eval_sdk_optimal/error_statistics.csvOther246 B —
eval_sdk_optimal/evaluation_metrics.csvOther394 B —
eval_sdk_optimal/f1_metrics.csvOther285 B —
eval_sdk_optimal/mar_metrics.csvOther473 B —
eval_sdk_optimal/per_class_metrics.csvOther265 B —
eval_sdk_optimal/performance_analysis.pngOther89.3 KB —
eval_sdk_optimal/pr_analysis_results.csvOther304 B —
eval_sdk_optimal/precision_metrics.csvOther277 B —
eval_sdk_optimal/prediction_samples.pngOther12.8 MB f270827ec7ce
eval_sdk_optimal/recall_metrics.csvOther279 B —
eval_sdk_optimal/results.pngOther43.1 KB —
evaluation/BoxF1_curve.pngOther125.6 KB 5e12129b00d2
evaluation/BoxPR_curve.pngOther102.2 KB df46df138aed
evaluation/BoxP_curve.pngOther104.4 KB 4fc51da2f253
evaluation/BoxR_curve.pngOther125.9 KB 5a510342e8fc
evaluation/MaskF1_curve.pngOther125.5 KB 06d999c3138d
evaluation/MaskPR_curve.pngOther111.9 KB 12f9a64180e5
evaluation/MaskP_curve.pngOther110.9 KB 036d732b9a80
evaluation/MaskR_curve.pngOther125.1 KB 58fc1fc4d4cd
evaluation/confusion_matrix.pngOther38.7 KB —
evaluation/error_analysis.pngOther185.7 KB dfe0d61785e3
evaluation/error_statistics.csvOther220 B —
evaluation/evaluation_metrics.csvOther395 B —
evaluation/f1_metrics.csvOther288 B —
evaluation/mar_metrics.csvOther473 B —
evaluation/per_class_metrics.csvOther249 B —
evaluation/performance_analysis.pngOther90.5 KB —
evaluation/pr_analysis_results.csvOther304 B —
evaluation/precision_metrics.csvOther279 B —
evaluation/prediction_samples.pngOther2.8 MB 98d19a62602e
evaluation/recall_metrics.csvOther279 B —
evaluation/results.pngOther44.3 KB —
events.out.tfevents.1777955468.f98b7cad610d.89638.0Other656.6 KB d6a33976b2e7
labels.jpgOther148.8 KB a097333017ac
results.csvOther46.8 KB —
tensorboard_server.logOther667 B —
train_batch0.jpgOther423.7 KB c506199a59d3
train_batch1.jpgOther489.2 KB 32a3fde56d0a
train_batch2.jpgOther363.4 KB f5b00b3393d5
.gitattributesRepository4.4 KB —

License and Download

License
apache-2.0
Access
Open weights, no gate
Download size
3.8 GB
Download from Black Bee Drones

Released by Black Bee Drones through its official repository on Hugging Face. Read the license.

Built From

  • Trained on (disclosed) blackbeedrones/sae-2026-hook

Evaluations

Each result is shown as reported, with the conditions its reporter stated. None is a SAVRN measurement. A comparison lines two results up only when their configuration, unit and setup are all stated and identical.

BenchmarkConditionsResultReported byRevisionDate
SAE Eletroquad 2026 - Hook Task Instance SegmentationMetric Box mAP@50Comparison conditions not established 0.9929 blackbeedrones
Publisher reported		 Evaluated revision not stated —
SAE Eletroquad 2026 - Hook Task Instance SegmentationMetric Box mAP@50-95Comparison conditions not established 0.964 blackbeedrones
Publisher reported		 Evaluated revision not stated —
SAE Eletroquad 2026 - Hook Task Instance SegmentationMetric Mask mAP@50Comparison conditions not established 0.9844 blackbeedrones
Publisher reported		 Evaluated revision not stated —
SAE Eletroquad 2026 - Hook Task Instance SegmentationMetric Mask mAP@50-95Comparison conditions not established 0.8535 blackbeedrones
Publisher reported		 Evaluated revision not stated —

Memory Requirements

PrecisionWeights in memory
As published3.8 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About sae-2026-hang-all-yolo26n-seg-v2-960

Can I use sae-2026-hang-all-yolo26n-seg-v2-960 commercially?

Yes. sae-2026-hang-all-yolo26n-seg-v2-960 is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

Similar Models

Model · Image segmentation

segformer-b2-finetuned-ade-512-512

NVIDIA

SegFormer model fine-tuned on ADE20k at resolution 512x512. It was introduced in the paper SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers by Xie et al. and first released in this repository. Disclaimer: The team releasing SegFormer did not write a model card for this model so this model card has been written by the Hugging Face team. SegFormer consists of a hierarchical Transformer encoder and a lightweight all-MLP decode head to achieve great results on semantic segmentation benchmarks such as ADE20K and Cityscapes. The hierarchical Transformer is first pre-trained on ImageNet-1k, after which a decode head is added and fine-tuned altogether on a…

Open weights other transformers
View model

Model · Image segmentation

segformer-b3-finetuned-ade-512-512

NVIDIA

SegFormer model fine-tuned on ADE20k at resolution 512x512. It was introduced in the paper SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers by Xie et al. and first released in this repository. Disclaimer: The team releasing SegFormer did not write a model card for this model so this model card has been written by the Hugging Face team. SegFormer consists of a hierarchical Transformer encoder and a lightweight all-MLP decode head to achieve great results on semantic segmentation benchmarks such as ADE20K and Cityscapes. The hierarchical Transformer is first pre-trained on ImageNet-1k, after which a decode head is added and fine-tuned altogether on a…

Open weights other transformers
View model

Model · Image segmentation

oneformer_ade20k_swin_large

SHI Labs

OneFormer model trained on the ADE20k dataset (large-sized version, Swin backbone). It was introduced in the paper OneFormer: One Transformer to Rule Universal Image Segmentation by Jain et al. and first released in this repository. OneFormer is the first multi-task universal image segmentation framework. It needs to be trained only once with a single universal architecture, a single model, and on a single dataset, to outperform existing specialized models across semantic, instance, and panoptic segmentation tasks. OneFormer uses a task token to condition the model on the task in focus, making the architecture task-guided for training, and task-dynamic for inference, all with a single…

Open weights mit transformers
View model

Model · Image segmentation

segformer-b1-finetuned-ade-512-512

NVIDIA

SegFormer model fine-tuned on ADE20k at resolution 512x512. It was introduced in the paper SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers by Xie et al. and first released in this repository. Disclaimer: The team releasing SegFormer did not write a model card for this model so this model card has been written by the Hugging Face team. SegFormer consists of a hierarchical Transformer encoder and a lightweight all-MLP decode head to achieve great results on semantic segmentation benchmarks such as ADE20K and Cityscapes. The hierarchical Transformer is first pre-trained on ImageNet-1k, after which a decode head is added and fine-tuned altogether on a…

Open weights other transformers
View model

Model · Image segmentation

segformer-b0-finetuned-ade-512-512

Joshua

https://huggingface.co/nvidia/segformer-b0-finetuned-ade-512-512 with ONNX weights to be compatible with Transformers.js. If you haven't already, you can install the Transformers.js JavaScript library from NPM using: Example: Image segmentation with Xenova/segformer-b0-finetuned-ade-512-512. You can visualize the outputs with: Note: Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using Optimum and structuring your repo like this one (with ONNX weights located in a subfolder named onnx).

Open weights transformers.js
View model

Model · Image segmentation

DelineateAnything

Mykola Lavreniuk

Delineate Anything v2 extends Delineate Anything into a globally representative, resolution-agnostic foundation model that scales agricultural field boundary detection to a planetary level from any imagery source. Trained on FBIS-73M, a massive 73-million-instance dataset spanning 61 countries with diverse imagery sources ranging from 0.25m to 10m resolution, built through a resolution-specific curation pipeline that solves the parcel-versus-field mismatch, Delineate Anything v2 sets a new state-of-the-art in global zero-shot delineation. It delivers a +103.3% relative gain in [email protected] over Delineate Anything while maintaining extreme efficiency, mapping all of Ukraine (603,000 km²) in 5.4…

Open weights agpl-3.0 ultralytics
View model