← kembali

kategori

Model serving

Serve trained machine learning models behind an API, with batching, scaling and versioning.

Perbandingan berdampingan

Model serving
AlatBintangLisensiKetentuanSelf-hostedBahasaTerbaruPush terakhir
Triton Inference Server11kBSD-3-ClauseOpen sourceYaPythonv2.73.02026-10-06
BentoML8.9kApache-2.0Open sourceYaPythonv1.4.392026-10-05
KServe6.1kApache-2.0Open sourceYaGov0.21.02026-10-06
MLServer900Apache-2.0Open sourceYaPython1.7.12026-10-05

4 alat

BentoML

The easiest way to serve AI apps and models - Build Model Inference APIs, Job queues, LLM apps, Multi-model pipelines, and more!

Bahasa
Python
Lisensi
Apache-2.0
Bintang
8.9k
Terbaru
v1.4.39
Push terakhir
2026-10-05
KServe

Standardized Distributed Generative and Predictive AI Inference Platform for Scalable, Multi-Framework Deployment on Kubernetes

Bahasa
Go
Lisensi
Apache-2.0
Bintang
6.1k
Terbaru
v0.21.0
Push terakhir
2026-10-06
MLServer

An inference server for your machine learning models, including support for multiple frameworks, multi-model serving and more

Bahasa
Python
Lisensi
Apache-2.0
Bintang
900
Terbaru
1.7.1
Push terakhir
2026-10-05

Apa yang digantikan alat-alat ini