Prebuilt containers for prediction and explanation

Vertex AI provides Docker container images that you run as prebuilt containers for serving predictions and explanations from trained model artifacts. These containers, which are organized by machine learning (ML) framework and framework version, provide HTTP prediction servers that you can use to serve predictions with minimal configuration. In many cases, using a prebuilt container is simpler than creating your own custom container for prediction.

This document lists the prebuilt containers for predictions and explanations, and it describes how to use them with model artifacts that you created using Vertex AI's custom training functionality or model artifacts that you created outside of Vertex AI.

Vertex AI supports each framework version based on a schedule to minimize security vulnerabilities. Review the Vertex AI framework support policy to understand the implications of the end-of-support and end-of-availability dates.

Available container images

Each of the following container images is available in several Artifact Registry repositories, which store data in various locations. You can use any of the URIs for an image when you perform custom training; each provides the same container image. If you use the Google Cloud console to create a Model resource, the Google Cloud console selects the URI that best matches the location where you are using Vertex AI in order to reduce latency.

TensorFlow

ML framework version Supported accelerators (and CUDA version, if applicable) End of patch and support date End of availability Supported images
2.12 CPU only June 30, 2024 June 30, 2025 Planned for June 30, 2023
2.12 GPU (CUDA 11.x) June 30, 2024 June 30, 2025 Planned for June 30, 2023
2.11 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-11:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-11:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-11:latest
2.11 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-11:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-11:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-11:latest
2.10 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-10:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-10:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-10:latest
2.10 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-10:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-10:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-10:latest
2.9 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-9:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-9:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-9:latest
2.9 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-9:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-9:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-9:latest
2.8 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-8:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-8:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-8:latest
2.8 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-8:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-8:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-8:latest
2.7 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-7:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-7:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-7:latest
2.7 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-7:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-7:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-7:latest
2.6 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-6:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-6:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-6:latest
2.6 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-6:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-6:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-6:latest
2.5 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-5:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-5:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-5:latest
2.5 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-5:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-5:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-5:latest
2.4 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-4:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-4:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-4:latest
2.4 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-4:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-4:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-4:latest
2.3 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-3:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-3:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-3:latest
2.3 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-3:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-3:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-3:latest
2.2 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-2:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-2:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-2:latest
2.2 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-2:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-2:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-2:latest
2.1 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-1:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-1:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-cpu.2-1:latest
2.1 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-2:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-2:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf2-gpu.2-2:latest
1.15 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf-cpu.1-15:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf-cpu.1-15:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf-cpu.1-15:latest
1.15 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/tf-gpu.1-15:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/tf-gpu.1-15:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/tf-gpu.1-15:latest

Optimized TensorFlow runtime

The following container images use the optimized TensorFlow runtime. For more information, see Use the optimized TensorFlow runtime.

ML framework version Supported accelerators (and CUDA version, if applicable) End of patch and support date End of availability Supported images
nightly CPU only Not applicable Not applicable
  • us-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.nightly:latest
  • europe-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.nightly:latest
  • asia-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.nightly:latest
nightly GPU (CUDA 11.x) Not applicable Not applicable
  • us-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.nightly:latest
  • europe-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.nightly:latest
  • asia-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.nightly:latest
2.12 CPU only May 15, 2024 May 15, 2025
  • us-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-12:latest
  • europe-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-12:latest
  • asia-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-12:latest
2.12 GPU (CUDA 11.x) May 15, 2024 May 15, 2025
  • us-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-12:latest
  • europe-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-12:latest
  • asia-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-12:latest
2.11 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-11:latest
  • europe-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-11:latest
  • asia-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-11:latest
2.11 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-11:latest
  • europe-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-11:latest
  • asia-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-11:latest
2.10 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-10:latest
  • europe-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-10:latest
  • asia-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-10:latest
2.10 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-10:latest
  • europe-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-10:latest
  • asia-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-10:latest
2.9 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-9:latest
  • europe-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-9:latest
  • asia-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-9:latest
2.9 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-9:latest
  • europe-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-9:latest
  • asia-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-9:latest
2.8 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-8:latest
  • europe-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-8:latest
  • asia-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-cpu.2-8:latest
2.8 GPU (CUDA 11.x) Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-8:latest
  • europe-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-8:latest
  • asia-docker.pkg.dev/vertex-ai-restricted/prediction/tf_opt-gpu.2-8:latest

PyTorch

ML framework version Supported accelerators (and CUDA version, if applicable) End of patch and support date End of availability Supported images
1.13 (Python 3.8) CPU only May 15, 2024 May 15, 2025
  • us-docker.pkg.dev/vertex-ai/prediction/pytorch-cpu.1-13:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/pytorch-cpu.1-13:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/pytorch-cpu.1-13:latest
1.13 (Python 3.8) GPU (CUDA 11.x) May 15, 2024 May 15, 2025
  • us-docker.pkg.dev/vertex-ai/prediction/pytorch-gpu.1-13:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/pytorch-gpu.1-13:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/pytorch-gpu.1-13:latest
1.12 CPU only May 15, 2024 May 15, 2025
  • us-docker.pkg.dev/vertex-ai/prediction/pytorch-cpu.1-12:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/pytorch-cpu.1-12:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/pytorch-cpu.1-12:latest
1.12 GPU (CUDA 11.x) May 15, 2024 May 15, 2025
  • us-docker.pkg.dev/vertex-ai/prediction/pytorch-gpu.1-12:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/pytorch-gpu.1-12:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/pytorch-gpu.1-12:latest
1.11 CPU only May 15, 2024 May 15, 2025
  • us-docker.pkg.dev/vertex-ai/prediction/pytorch-cpu.1-11:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/pytorch-cpu.1-11:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/pytorch-cpu.1-11:latest
1.11 GPU (CUDA 11.x) May 15, 2024 May 15, 2025
  • us-docker.pkg.dev/vertex-ai/prediction/pytorch-gpu.1-11:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/pytorch-gpu.1-11:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/pytorch-gpu.1-11:latest

scikit-learn

ML framework version Supported accelerators (and CUDA version, if applicable) End of patch and support date End of availability Supported images
1.2 (Python 3.10) CPU only June 30, 2024 June 30, 2025 Planned for June 30, 2023
1.0 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.1-0:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.1-0:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.1-0:latest
0.24 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.0-24:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.0-24:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.0-24:latest
0.23 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.0-23:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.0-23:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.0-23:latest
0.22 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.0-22:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.0-22:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.0-22:latest
0.20 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.0-20:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.0-20:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/sklearn-cpu.0-20:latest

XGBoost

ML framework version Supported accelerators (and CUDA version, if applicable) End of patch and support date End of availability Supported images
1.7 (Python 3.10) CPU only June 30, 2024 June 30, 2025 Planned for June 30, 2023
1.6 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-6:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-6:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-6:latest
1.5 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-5:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-5:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-5:latest
1.4 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-4:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-4:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-4:latest
1.3 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-3:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-3:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-3:latest
1.2 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-2:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-2:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-2:latest
1.1 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-1:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-1:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.1-1:latest
0.90 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.0-90:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.0-90:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.0-90:latest
0.82 CPU only Nov 15, 2023 Nov 15, 2024
  • us-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.0-82:latest
  • europe-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.0-82:latest
  • asia-docker.pkg.dev/vertex-ai/prediction/xgboost-cpu.0-82:latest

Use a prebuilt container

You can specify a prebuilt container for prediction when you create a custom TrainingPipeline resource that uploads a Model or when you import model artifacts as a Model.

To use one of these prebuilt containers, you must save your model as one or more model artifacts that comply with the requirements of the prebuilt container. For more information, see Export model artifacts for prediction.

The following notebooks demonstrate how to use a prebuilt container to serve predictions.

What do you want to do? Notebook
Train and serve a TensorFlow model using a prebuilt container Custom training and online prediction
Serve a PyTorch model using a prebuilt container Serving PyTorch image models with prebuilt containers on Vertex AI
Serve a Stable Diffusion model using a prebuilt container Deploy and host a Stable Diffusion model on Vertex AI

What's next