Managed AWS service for building, training and deploying machine learning models. Um produto fechado de Amazon Web Services.
| Ferramenta | Substituição | Estrelas | Licença | Termos | Auto-hospedado | Linguagem | Última release | Último push |
|---|---|---|---|---|---|---|---|---|
| Triton Inference Server | Parcial | 11k | BSD-3-Clause | Código aberto | Sim | Python | v2.73.0 | 2026-10-06 |
| Metaflow | Parcial | 10k | Apache-2.0 | Código aberto | Sim | Python | 2.19.39 | 2026-10-06 |
| BentoML | Parcial | 8.9k | Apache-2.0 | Código aberto | Sim | Python | v1.4.39 | 2026-10-05 |
| KServe | Parcial | 6.1k | Apache-2.0 | Código aberto | Sim | Go | v0.21.0 | 2026-10-06 |
| ZenML | Parcial | 5.6k | Apache-2.0 | Código aberto | Sim | Python | 0.97.0 | 2026-10-06 |
| Kubeflow Pipelines | Parcial | 4.2k | Apache-2.0 | Código aberto | Sim | Go | 2.17.2 | 2026-10-06 |
| MLRun | Parcial | 1.7k | Apache-2.0 | Código aberto | Sim | Python | v1.12.0 | 2026-10-06 |
| MLServer | Parcial | 900 | Apache-2.0 | Código aberto | Sim | Python | 1.7.1 | 2026-10-05 |
8 alternativas
The Triton Inference Server provides an optimized cloud and edge inferencing solution.
Serves TensorRT, PyTorch, ONNX and other models over HTTP and gRPC, with dynamic batching.
Build, Manage and Deploy AI/ML Systems
Python flows that scale from a laptop to Kubernetes or AWS Batch, with versioned artefacts.
The easiest way to serve AI apps and models - Build Model Inference APIs, Job queues, LLM apps, Multi-model pipelines, and more!
Packages models as API services and container images to run anywhere.
Standardized Distributed Generative and Predictive AI Inference Platform for Scalable, Multi-Framework Deployment on Kubernetes
Kubernetes resources for serving predictive and generative models with autoscaling.
ZenML 🙏: One AI Platform from Pipelines to Agents. https://zenml.io.
Portable pipelines that run on your own orchestrators and clouds, with a model registry.
Machine Learning Pipelines for Kubeflow
Pipelines of containerised ML steps on Kubernetes.
MLRun is an open source MLOps platform for quickly building and managing continuous ML applications across their lifecycle. MLRun integrates into your development and CI/CD environment and automates the delivery of production data, ML pipelines, and online applications.
Pipelines, model serving and monitoring on Kubernetes.
An inference server for your machine learning models, including support for multiple frameworks, multi-model serving and more
A Python inference server speaking the V2 inference protocol.