diff --git a/projects/TPVFormer/README.md b/projects/TPVFormer/README.md
index 9a0681bf88..fcc06ec679 100644
--- a/projects/TPVFormer/README.md
+++ b/projects/TPVFormer/README.md
@@ -24,7 +24,7 @@ We implement TPVFormer and provide the results and checkpoints on nuScenes datas
 
 In MMDetection3D's root directory, run the following command to train the model:
 
-1. Downloads the [pretrained backbone weights](<>) to checkpoints/
+1. Downloads the [pretrained backbone weights](https://download.openmmlab.com/mmdetection3d/v1.1.0_models/tpvformer/tpvformer_8xb1-2x_nus-seg/tpvformer_pretrained_fcos3d_r101_dcn.pth) to checkpoints/
 
 2. For example, to train TPVFormer on 8 GPUs, please use
 
diff --git a/projects/TPVFormer/configs/tpvformer_8xb1-2x_nus-seg.py b/projects/TPVFormer/configs/tpvformer_8xb1-2x_nus-seg.py
index 7861c6f13e..70021d8ff5 100644
--- a/projects/TPVFormer/configs/tpvformer_8xb1-2x_nus-seg.py
+++ b/projects/TPVFormer/configs/tpvformer_8xb1-2x_nus-seg.py
@@ -258,7 +258,7 @@
         stage_with_dcn=(False, False, True, True),
         init_cfg=dict(
             type='Pretrained',
-            checkpoint='checkpoints/tpvformer_r101_dcn_fcos3d_pretrain.pth',
+            checkpoint='checkpoints/tpvformer_pretrained_fcos3d_r101_dcn.pth',
             prefix='backbone.')),
     neck=dict(
         type='mmdet.FPN',
@@ -270,7 +270,7 @@
         relu_before_extra_convs=True,
         init_cfg=dict(
             type='Pretrained',
-            checkpoint='checkpoints/tpvformer_r101_dcn_fcos3d_pretrain.pth',
+            checkpoint='checkpoints/tpvformer_pretrained_fcos3d_r101_dcn.pth',
             prefix='neck.')),
     encoder=dict(
         type='TPVFormerEncoder',