[BUG] tf.config.list_physical_devices("GPU") cannot find available GPUs

Description

The tests failed due to the GPU check using tf.config.list_physical_devices("GPU") on NVIDIA A800. However, torch.cuda can successfully find the GPU.

>>> import recommenders.utils.gpu_utils as gpu_utils
>>> gpu_utils.get_gpu_info()
                                                         
[{'device_name': 'NVIDIA A800-SXM4-80GB', 'total_memory': 81182.0, 'free_memory': 80757.1875}]
>>> import tensorflow as tf
2026-09-16 10:03:41.760566: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-09-16 10:03:41.762633: I external/local_tsl/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used.
2026-09-16 10:03:41.794912: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-09-16 10:03:41.794946: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-09-16 10:03:41.795979: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-09-16 10:03:41.801656: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-09-16 10:03:44.223224: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
>>> tf.config.list_physical_devices("GPU")
2026-09-16 10:04:05.156264: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-09-16 10:04:05.157680: W tensorflow/core/common_runtime/gpu/gpu_device.cc:2256] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
[]

In which platform does it happen?

• NVIDIA A800-SXM4-80G with
  • CUDA 13.4
  • CUDA driver 615.71.09
  • Python 3.10.20
  • TensorFlow 2.15.1

How do we replicate the issue?

Expected behavior (i.e. solution)

Willingness to contribute

• Yes, I can contribute for this issue independently.
• Yes, I can contribute for this issue with guidance from Recommenders community.
• No, I cannot contribute at this time.

Other Comments