---

The easiest way to build Machine Learning APIs

Multi-framework / High-performance / Easy to learn / Production ready

What does BentoML do?

• Package models trained with any ML framework and reproduce them for model serving in production

• Package once and deploy anywhere for real-time API serving or offline batch serving

• High-Performance API model server with adaptive micro-batching support

• Central storage hub with Web UI and APIs for managing and accessing packaged models

• Modular and flexible design allowing advanced users to easily customize

BentoML is a framework for serving, managing and deploying machine learning models. It is aiming to bridge the gap between Data Science and DevOps, and enable data science teams to continuesly deliver prediction services to production.

💻 Get started with BentoML: Quickstart Guide | Quickstart on Google Colab

👩‍💻 Star/Watch/Fork the BentoML Github Repository.

👉 Join the community: Bentoml Slack Channel and the Discussions on Github.

---

• Getting Started
  • Run on Google Colab
  • Run notebook locally
  • Hello World
  • Save prediction service for distribution
  • Model Serving via REST API
  • Launch inference job from CLI
  • Containerize Model API Server
  • Deployment Options
  • Distribute BentoML packaged model as a PyPI library
  • Learning more about BentoML
• Core Concepts
  • Creating BentoService
  • Defining Service Environment
  • Packaging Model Artifacts
  • API Function and Adapters
  • Saving BentoService
  • Model Serving
  • Model Management
  • Retrieving BentoServices
• Frameworks
  • Scikit-Learn
  • PyTorch
  • Tensorflow 2.0
  • Tensorflow Keras
  • FastAI
  • FastAI v2
  • XGBoost
  • LightGBM
  • FastText
  • Statsmodels
  • H2O
  • CoreML
  • ONNX
  • Spacy
  • Transformers
  • Statsmodels
• Advanced Guides
  • Offline Batch Serving
  • Monitoring with Prometheus
  • Request Logging
  • Understanding BentoML adaptive micro batching
  • 1. The overall architecture of BentoML’s micro-batching server
  • 2. parameter tuning best practices & recommendations
  • 3. How to implement batch mode for custom input adapters
  • 4. Comparison
  • Adding Custom Model Artifact
  • Customizing InputAdapter
  • Deploy yatai server behind NGINX
  • Using Helm to install YataiService
  • 1. Configuration
  • 2. Deploying
• Deployment Guides
  • Deploying to AWS Lambda
  • Deploying to AWS SageMaker
  • Deploying to Azure Functions
  • Deploying to Clipper Cluster
  • Deploying to AWS ECS(Elastic Container Service)
  • Deploying to Google Cloud Run
  • Deploying to Azure Container Instance
  • Deploying to Kubernetes Cluster
  • Deploying to KNative
  • Deploying to Kubeflow
  • Deploying to KFServing
  • Deploying to Heroku
  • Deploying to SQL Server Machine Learning Services
• Example Projects
  • FastAI
  • Scikit-Learn
  • PyTorch
  • Tensorflow Keras
  • Tensorflow 2.0
  • XGBoost
  • LightGBM
  • H2O
  • FastText
  • ONNX
  • SQL Server Machine Learning Services
  • Statsmodels
• API Reference
  • BentoService API
  • API InputAdapters
  • API OutputAdapters
  • Adapter Roadmap
  • Frameworks
  • Base Artifact
• CLI Reference
  • bentoml
• Frequently Asked Questions
  • Why BentoML?
  • How does BentoML compare to Tensorflow-serving?
  • How does BentoML compare to Clipper?
  • How does BentoML compare to AWS SageMaker?
  • How does BentoML compare to MLFlow?
  • Does BentoML do horizontal scaling?
  • How does BentoML compare with Cortex?
  • How does BentoML compare to Seldon?
  • Is there a plan for R support?