¿Qué estoy haciendo mal? Teslat T4 CUDA yolov5

Hola, parece que no puedo lograr que la Detección de Objetos del Proyecto AI habilite mi Tesla T4 recién instalado. ¿Alguna idea de lo que estoy haciendo mal? YOLOv5 no tiene la opción de habilitar la GPU y solo se muestra la CPU.

17:57:27:System:           Windows
17:57:27:Operating System: Windows (Microsoft Windows 10.0.19045)
17:57:27:CPUs:             Intel(R) Core(TM) i5-3570 CPU @ 3.40GHz (Intel)
17:57:27:                  1 CPU x 4 cores. 4 logical processors (x64)
17:57:27:GPU (Primary):    Tesla P4 (8 GiB) (NVIDIA)
17:57:27:                  Driver: 551.61, CUDA: 12.4 (up to: 12.4), Compute: 6.1, cuDNN: 8.9
17:57:27:System RAM:       16 GiB
17:57:27:Platform:         Windows
17:57:27:BuildConfig:      Release
17:57:27:Execution Env:    Native
17:57:27:Runtime Env:      Production
17:57:27:Runtimes installed:
17:57:27:  .NET runtime:     7.0.10
17:57:27:  .NET SDK:         Not found
17:57:27:  Default Python:   Not found
17:57:27:  Go:               Not found
17:57:27:  NodeJS:           Not found
17:57:27:App DataDir:      C:\ProgramData\CodeProject\AI
17:57:27:Video adapter info:
17:57:27:  Intel(R) HD Graphics:
17:57:27:    Driver Version     10.18.10.4425
17:57:27:    Video Processor    Intel(R) HD Graphics Family
17:57:27:  NVIDIA Tesla P4:
17:57:27:    Driver Version     31.0.15.5161
17:57:27:    Video Processor
17:57:27:STARTING CODEPROJECT.AI SERVER
17:57:27:RUNTIMES_PATH             = C:\Program Files\CodeProject\AI\runtimes
17:57:27:PREINSTALLED_MODULES_PATH = C:\Program Files\CodeProject\AI\preinstalled-modules
17:57:27:MODULES_PATH              = C:\Program Files\CodeProject\AI\modules
17:57:27:PYTHON_PATH               = \bin\windows\%PYTHON_NAME%\venv\Scripts\python
17:57:27:Data Dir                  = C:\ProgramData\CodeProject\AI
17:57:27:Server version:   2.5.6
17:57:33:Server: This is the latest version
18:10:11:Preparing to install module 'ObjectDetectionYOLOv5-3.1'
18:10:11:Downloading module 'ObjectDetectionYOLOv5-3.1'
18:10:11:Installing module 'ObjectDetectionYOLOv5-3.1'
18:10:11:ObjectDetectionYOLOv5-3.1:              Installing CodeProject.AI Analysis Module                
18:10:11:ObjectDetectionYOLOv5-3.1: ======================================================================
18:10:11:ObjectDetectionYOLOv5-3.1:                    CodeProject.AI Installer                           
18:10:11:ObjectDetectionYOLOv5-3.1: ======================================================================
18:10:12:ObjectDetectionYOLOv5-3.1: 140.9Gb of 227Gb available on OS
18:10:12:ObjectDetectionYOLOv5-3.1: General CodeProject.AI setup                                          
18:10:12:ObjectDetectionYOLOv5-3.1: Creating Directories...Done
18:10:12:ObjectDetectionYOLOv5-3.1: GPU support                                                           
18:10:12:ObjectDetectionYOLOv5-3.1: CUDA Present...Yes (CUDA 11.8, cuDNN 8.9)
18:10:12:ObjectDetectionYOLOv5-3.1: ROCm Present...No
18:10:16:ObjectDetectionYOLOv5-3.1: Reading ObjectDetectionYOLOv5-3.1 settings.......Done
18:10:16:ObjectDetectionYOLOv5-3.1: Installing module Object Detection (YOLOv5 3.1) 1.9.1                 
18:10:16:ObjectDetectionYOLOv5-3.1: Installing Python 3.7
18:10:16:ObjectDetectionYOLOv5-3.1: Python 3.7 is already installed
18:10:26:ObjectDetectionYOLOv5-3.1: Creating Virtual Environment (Local)...Done
18:10:26:ObjectDetectionYOLOv5-3.1: Confirming we have Python 3.7 in our virtual environment...present
18:10:27:ObjectDetectionYOLOv5-3.1: Downloading Standard YOLOv5 models...already exists...Expanding...Done.
18:10:27:ObjectDetectionYOLOv5-3.1: Copying contents of models-yolo5-31-pt.zip to assets...done
18:10:28:ObjectDetectionYOLOv5-3.1: Downloading Custom YOLOv5 models...already exists...Expanding...Done.
18:10:28:ObjectDetectionYOLOv5-3.1: Copying contents of custom-models-yolo5-31-pt.zip to custom-models...done
18:10:28:ObjectDetectionYOLOv5-3.1: Installing Python packages for Object Detection (YOLOv5 3.1)
18:10:28:ObjectDetectionYOLOv5-3.1: [0;Installing GPU-enabled libraries: If available
18:10:29:ObjectDetectionYOLOv5-3.1: Ensuring Python package manager (pip) is installed...Done
18:10:41:Response timeout. Try increasing the timeout value
18:10:45:ObjectDetectionYOLOv5-3.1: Ensuring Python package manager (pip) is up to date...Done
18:10:45:ObjectDetectionYOLOv5-3.1: Python packages specified by requirements.windows.cuda11.txt
18:10:56:ObjectDetectionYOLOv5-3.1:   - Installing OpenCV, the Open source Computer Vision library...(checked) Done
18:11:02:ObjectDetectionYOLOv5-3.1:   - Installing the Cython compiler for C extensions for the Python language....(checked) Done
18:11:06:ObjectDetectionYOLOv5-3.1:   - Installing Pillow, a Python Image Library...(checked) Done
18:11:26:ObjectDetectionYOLOv5-3.1:   - Installing SciPy, a library for mathematics, science, and engineering...(checked) Done
18:11:29:ObjectDetectionYOLOv5-3.1:   - Installing PyYAML, a library for reading configuration files...(checked) Done
18:11:32:ObjectDetectionYOLOv5-3.1:   - Installing TDQM, the Fast, Extensible Progress Meter...(checked) Done
18:11:53:ObjectDetectionYOLOv5-3.1:   - Installing Matplotlib, the Python plotting package...(checked) Done
18:13:57:ObjectDetectionYOLOv5-3.1:   - Installing PyTorch, an open source machine learning framework...(checked) Done
18:16:28:ObjectDetectionYOLOv5-3.1:   - Installing TorchVision, for working with computer vision models...(checked) Done
18:16:28:ObjectDetectionYOLOv5-3.1: Installing Python packages for the CodeProject.AI Server SDK
18:16:29:ObjectDetectionYOLOv5-3.1: Ensuring Python package manager (pip) is installed...Done
18:16:32:ObjectDetectionYOLOv5-3.1: Ensuring Python package manager (pip) is up to date...Done
18:16:32:ObjectDetectionYOLOv5-3.1: Python packages specified by requirements.txt
18:16:33:ObjectDetectionYOLOv5-3.1:   - Installing Pillow, a Python Image Library...Already installed
18:16:34:ObjectDetectionYOLOv5-3.1:   - Installing Charset normalizer...Already installed
18:16:41:ObjectDetectionYOLOv5-3.1:   - Installing aiohttp, the Async IO HTTP library...(checked) Done
18:16:45:ObjectDetectionYOLOv5-3.1:   - Installing aiofiles, the Async IO Files library...(checked) Done
18:16:48:ObjectDetectionYOLOv5-3.1:   - Installing py-cpuinfo to allow us to query CPU info...(checked) Done
18:16:49:ObjectDetectionYOLOv5-3.1:   - Installing Requests, the HTTP library...Already installed
18:17:04:ObjectDetectionYOLOv5-3.1: Self test: Self-test passed
18:17:04:ObjectDetectionYOLOv5-3.1: Module setup time 00:06:51.89
18:17:04:ObjectDetectionYOLOv5-3.1: Setup complete                                                        
18:17:04:ObjectDetectionYOLOv5-3.1: Total setup time 00:06:52.92
18:17:04:Module ObjectDetectionYOLOv5-3.1 installed successfully.
18:17:04:Module ObjectDetectionYOLOv5-3.1 not configured to AutoStart.
18:17:04:Installer exited with code 0
18:20:21:Update ObjectDetectionYOLOv5-3.1. Setting AutoStart=true
18:20:21:Restarting Object Detection (YOLOv5 3.1) to apply settings change
18:20:21:
18:20:21:Module 'Object Detection (YOLOv5 3.1)' 1.9.1 (ID: ObjectDetectionYOLOv5-3.1)
18:20:21:Valid:         True
18:20:21:Module Path:   <root>\modules\ObjectDetectionYOLOv5-3.1
18:20:21:AutoStart:     True
18:20:21:Queue:         objectdetection_queue
18:20:21:Runtime:       python3.7
18:20:21:Runtime Loc:   Local
18:20:21:FilePath:      detect_adapter.py
18:20:21:Pre installed: False
18:20:21:Start pause:   1 sec
18:20:21:Parallelism:   0
18:20:21:LogVerbosity:
18:20:21:Platforms:     all,!macos-arm64
18:20:21:GPU Libraries: installed if available
18:20:21:GPU Enabled:   enabled
18:20:21:Accelerator:
18:20:21:Half Precis.:  enable
18:20:21:Environment Variables
18:20:21:APPDIR         = <root>\modules\ObjectDetectionYOLOv5-3.1
18:20:21:DATA_DIR       = C:\ProgramData\CodeProject\AI
18:20:21:MODE           = MEDIUM
18:20:21:MODELS_DIR     = <root>\modules\ObjectDetectionYOLOv5-3.1\assets
18:20:21:PROFILE        = desktop_gpu
18:20:21:TEMP_PATH      = <root>\modules\ObjectDetectionYOLOv5-3.1\tempstore
18:20:21:USE_CUDA       = True
18:20:21:YOLOv5_VERBOSE = false
18:20:21:
18:20:21:Started Object Detection (YOLOv5 3.1) module

Module 'Object Detection (YOLOv5 3.1)' 1.9.1 (ID: ObjectDetectionYOLOv5-3.1)
Valid:         True
Module Path:   <root>\modules\ObjectDetectionYOLOv5-3.1
AutoStart:     True
Queue:         objectdetection_queue
Runtime:       python3.7
Runtime Loc:   Local
FilePath:      detect_adapter.py
Pre installed: False
Start pause:   1 sec
Parallelism:   0
LogVerbosity:  
Platforms:     all,!macos-arm64
GPU Libraries: installed if available
GPU Enabled:   enabled
Accelerator:   
Half Precis.:  enable
Environment Variables
   APPDIR         = <root>\modules\ObjectDetectionYOLOv5-3.1
   DATA_DIR       = C:\ProgramData\CodeProject\AI
   MODE           = MEDIUM
   MODELS_DIR     = <root>\modules\ObjectDetectionYOLOv5-3.1\assets
   PROFILE        = desktop_gpu
   TEMP_PATH      = <root>\modules\ObjectDetectionYOLOv5-3.1\tempstore
   USE_CUDA       = True
   YOLOv5_VERBOSE = false
Status Data:  {
  "inferenceDevice": "GPU",
  "inferenceLibrary": "CUDA",
  "canUseGPU": "true",
  "successfulInferences": 0,
  "failedInferences": 0,
  "numInferences": 0,
  "averageInferenceMs": 0
}
Started:      11 Mar 2024 6:20:21 PM Pacific Standard Time
LastSeen:     11 Mar 2024 6:26:41 PM Pacific Standard Time
Status:       Started
Requests:     0 (includes status calls)

Lo que he probado:

He reinstalado todos los controladores y software necesarios.