Pretrained ConvNeXtV2

NVIDIA

Model

NVIDIA

Pretrained ConvNeXtV2

Pretrained ConvNextv2 backbone models to facilitate transfer learning for commercially viable models.

3 Versions

07/17/2025 12:13 AM UTC178.99 MBBatch Size: 1GPU: V100

TAO Toolkit API

Key	Value
sha256_digest	{'convnextv2_nano.pth': 'a222f5335a58687a09babd36ff83ca1c8d475c7ae0535549d022b01ef03f5581'}
endpoints	['mae', 'rtdetr']
format_version	0.0.1
model_format	pth
backbone	true
tao_version	['>=6.0.0']
trainable	true

07/17/2025 12:13 AM UTC327.94 MBBatch Size: 1GPU: V100

TAO Toolkit API

Key	Value
sha256_digest	{'convnextv2_tiny.pth': '859836438bd007cd2b1b5c5e89761db71df7b70e4d290327da984b4bfa5ac037'}
endpoints	['mae', 'rtdetr']
format_version	0.0.1
model_format	pth
backbone	true
tao_version	['>=6.0.0']
trainable	true

07/17/2025 12:13 AM UTC2.21 GBBatch Size: 1GPU: V100

TAO Toolkit API

Key	Value
sha256_digest	{'convnextv2_large.pth': '90cbef2864117f0cd0ffb399922c5e1a4203724769288a7f92836cba7bdcb272'}
endpoints	['mae', 'rtdetr']
format_version	0.0.1
model_format	pth
backbone	true
tao_version	['>=6.0.0']
trainable	true