OpenMMLab
lab · ChinaScores
1 product on the map — 1 open.
Openness
5 high confidence- license
- Apache-2.0 (OSI; code)
- source
- public (github.com/open-mmlab/mmdetection)
- core features withheld
- no — OpenMMLab research toolbox; nothing sold
The code is Apache-2.0 and builds from the public repository, and OpenMMLab sells nothing around it. The last tagged release was v3.3.0 in January 2024, and the repository has not been pushed to since August 2024.
- https://raw.githubusercontent.com/open-mmlab/mmdetection/main/LICENSE recorded 2026-11-12
LICENSE is the Apache License, Version 2.0, under "Copyright 2018-2023 OpenMMLab. All rights reserved."
- https://raw.githubusercontent.com/open-mmlab/mmdetection/main/README.md recorded 2026-11-12
README: "MMDetection is an open source object detection toolbox based on PyTorch. It is a part of the OpenMMLab project."; "This project is released under the Apache 2.0 license."; no paid tier mentioned.
- https://ungh.cc/repos/open-mmlab/mmdetection recorded 2026-11-12
Repository record for open-mmlab/mmdetection: not archived, last pushed 2024-08-21.
- https://ungh.cc/repos/open-mmlab/mmdetection/releases recorded 2026-11-12
Latest release v3.3.0, published 2024-01-05.
Adoption
3 high confidencePyPI downloads of the mmdet package over the trailing 30 days. A download count includes CI jobs and mirrors as well as people, so it measures installs rather than users.
- https://pypistats.org/api/packages/mmdet/recent recorded 2026-11-12
last_month 122,001 downloads for mmdet (last_week 30,460, last_day 4,760).
Capability
4 medium confidenceMMDetection runs detectors that take the object names as text at inference, so a new label set needs no training run, as Ultralytics does with YOLO-World and YOLOE. Most of its model zoo is still trained on fixed benchmark classes.
- https://raw.githubusercontent.com/open-mmlab/mmdetection/main/configs/mm_grounding_dino/README.md recorded 2026-11-12
MM-Grounding-DINO README: zero-shot COCO, LVIS, ODinW and referring-expression results with downloadable models; MM-GDINO-T 50.6 zero-shot COCO mAP with O365, GoldG and V3Det pretraining.
- https://raw.githubusercontent.com/open-mmlab/mmdetection/main/README.md recorded 2026-11-12
README What's New: v3.3.0 brings MM-Grounding-DINO, "an open source replication version of Grounding DINO" for open vocabulary object detection and phrase grounding; the model list includes GLIP and Detic.