diff --git a/.dev_scripts/benchmark_filter.py b/.dev_scripts/benchmark_filter.py
index 8d9d39e8f89709229dec8d6b609c1519dd3ff291..2097394a1f295bedaa488ae41d88fb6ee256ef16 100644
--- a/.dev_scripts/benchmark_filter.py
+++ b/.dev_scripts/benchmark_filter.py
@@ -30,7 +30,7 @@ basic_arch_root = [
     'cascade_rcnn', 'double_heads', 'fcos', 'foveabox', 'free_anchor',
     'grid_rcnn', 'guided_anchoring', 'htc', 'libra_rcnn', 'atss', 'mask_rcnn',
     'ms_rcnn', 'nas_fpn', 'reppoints', 'retinanet', 'ssd', 'gn', 'ghm', 'fsaf',
-    'point_rend', 'nas_fcos', 'pisa', 'dynamic_rcnn'
+    'point_rend', 'nas_fcos', 'pisa', 'dynamic_rcnn', 'gfl', 'sabl', 'paa', 'yolo'
 ]
 
 datasets_root = ['wider_face', 'pascal_voc', 'cityscapes', 'mask_rcnn']
@@ -90,7 +90,11 @@ benchmark_pool = [
     'configs/pascal_voc/ssd300_voc0712.py',
     'configs/nas_fpn/retinanet_r50_nasfpn_crop640_50e_coco.py',
     'configs/mask_rcnn/mask_rcnn_r50_caffe_fpn_mstrain-poly_1x_coco.py',
-    'configs/gn+ws/mask_rcnn_r50_fpn_gn_ws-all_2x_coco.py'
+    'configs/gn+ws/mask_rcnn_r50_fpn_gn_ws-all_2x_coco.py',
+    'configs/gfl/gfl_r50_fpn_1x_coco.py',
+    'configs/paa/paa_r50_fpn_1x_coco.py',
+    'configs/sabl/sabl_retinanet_r50_fpn_1x_coco.py',
+    'configs/yolo/yolov3_d53_320_273e_coco.py'
 ]
 
 
diff --git a/docs/changelog.md b/docs/changelog.md
index 4cc79f9bfaffe38ee66718738c81e331ff0a082a..1ca2406dee9a3c03566ffe789a790b461d8d189b 100644
--- a/docs/changelog.md
+++ b/docs/changelog.md
@@ -4,8 +4,10 @@
 
 **Highlights**
 - Refactor common issues and add trouble shooting (#3530)
-- Support new methods [Yolact](https://arxiv.org/abs/1904.02689), [PAA Assign](https://arxiv.org/abs/2007.08103)
+- Support new methods [SABL](https://arxiv.org/abs/1912.04260), [YOLOv3](https://arxiv.org/abs/1804.02767), and [PAA Assign](https://arxiv.org/abs/2007.08103)
 - Support Batch Inference
+- Start to publish `mmdet` package to PyPI since v2.4.0
+- Switch model zoo to aliyun
 
 **Bug Fixes**
 - Fix default mean/std for onnx (#3491)
@@ -22,14 +24,18 @@
 - Fix import of fuse conv bn in benchmark.py (#3606)
 - Fix webcam demo (#3634)
 - Fix typo and itemize issues in tutorial (#3658)
+- Fix error in distributed training when some levels of FPN are not assigned with bounding boxes (#3670)
 
 **New Features**
-- Support Cutout (#3521)
+- Support Cutout augmentation (#3521)
 - Support evaluation on multiple datasets through ConcatDataset (#3522)
 - Support [PAA assign](https://arxiv.org/abs/2007.08103) #(3547)
 - Support eval metric with pickle results (#3607)
-- Support [YOLACT](https://arxiv.org/abs/1904.02689) (#3456)
+- Support [YOLOv3](https://arxiv.org/abs/1804.02767) (#3083)
+- Support [SABL](https://arxiv.org/abs/1912.04260) (#3603)
 - Support Batch Inference (#3564)
+- Support to publish to Pypi in github-action (#3510)
+- Support RandomFlip with horizontal/vertical/diagonal direction (#3608)
 
 **Improvements**
 - Refactor common issues in documentation (#3530)
@@ -41,6 +47,7 @@
 - Using config's loading in inference_detector (#3611)
 - Add atss 101 (#3639)
 - Support custom imports (#3641)
+- Update urls to download.openmmlab.com (#3665)
 
 
 ### v2.3.0 (5/8/2020)
diff --git a/mmdet/__init__.py b/mmdet/__init__.py
index 2a0cd7e2feea22cc32f5fbc9490d5868f0744880..9cc009b5984394752e7a27c467b70a21ca73bd92 100644
--- a/mmdet/__init__.py
+++ b/mmdet/__init__.py
@@ -16,7 +16,7 @@ def digit_version(version_str):
 
 
 mmcv_minimum_version = '1.1.1'
-mmcv_maximum_version = '1.1.1'
+mmcv_maximum_version = '1.2'
 mmcv_version = digit_version(mmcv.__version__)