Is there anyway to update older tao toolkit docker container tag 4.0.0-tf1.15.5 to support newer GPU like 5070?

Accelerated Computing Intelligent Video Analytics TAO Toolkit
1

Currently, I’ve been taking the course “Synthetic Data Generation for Perception Model Training in Isaac Sim” in Robotic Fundamental Learning Path.

I try to run detectnet_v2 training example from the course inside docker container tag 4.0.0-tf1.15.5 in Jupyter notebook and I get this error:

`WARNING: Detected NVIDIA NVIDIA GeForce RTX 5070 GPU, which is not yet supported in this version of the container
ERROR: No supported GPU(s) detected to run this container`

Is there a quick way to update 4.0.0-tf1.15.5 docker to support newer GPU like 5070?

2

Could you try nvcr.io/nvidia/tao/tao-toolkit:5.0.0-tf1.15.5? If not work, please use an older version of GPU to run 4.0.0-tf1.15.5 or 4.0.1-tf1.15.5..

3

I try it now in 5.0.0-tf1.15.5, and it is still not supported. The same error occurs:

Screenshot from 2025-10-26 08-10-14
Screenshot from 2025-10-26 08-10-14953×278 36.9 KB

Anyway, Thank you for your reply.

4

Can I use 6.0.0-deploy for fine-tune training model? I see that it still includes detectnet_v2 as shown in Tao Toolkit document

Screenshot from 2025-10-26 08-43-27
Screenshot from 2025-10-26 08-43-27744×852 62.8 KB

5

The 6.0.0-deploy is a container for deployment instead of training.