# pytorch/serve

Serve, optimize and scale PyTorch models in production

Repository: https://github.com/pytorch/serve
Canonical: https://ross.abutalabs.com/products/pytorch-serve
Homepage: https://pytorch.org/serve/
Language: Java
License: Apache-2.0
License Family: permissive
Topics: pytorch, machine-learning, mlops, serving, docker, kubernetes, optimization, cpu, gpu, metrics, deep-learning
Archived: true
Last push: 2025-08-06T19:17:08+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 35, release rhythm 8, longevity 100
- inputs: {"age_days": 2526, "days_push": 392, "days_rel": 702, "gap_med": null, "n_releases_24m": 1}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4346, forks 880 (observed 2026-08-28T04:08:46.198715+00:00)

## What it is
TorchServe is a flexible, production-ready model server for serving, optimizing, and scaling PyTorch models over HTTP with support for CPU, GPU, Docker, and Kubernetes. It packages models via torch-model-archiver and exposes prediction APIs with metrics and token-based authorization.

## Use cases
- serve pytorch models in production
- deploy a trained pytorch model behind a REST API
- scale model inference on kubernetes
- serve bert model for predictions over http
- run pytorch inference on gpu with docker
- monitor model serving metrics
- package pytorch models into a mar file

## When to choose
- you need a battle-tested HTTP inference server specifically for PyTorch models
- you want built-in model versioning, batching, metrics, and Kubernetes integration
- you need CPU and GPU serving with Docker images out of the box

## When to avoid
- you need active maintenance and security patches - the project is in limited maintenance mode
- you serve non-PyTorch frameworks like TensorFlow or ONNX exclusively
- you want a lightweight custom inference service you fully control
- you need cutting-edge LLM serving features better covered by vLLM or similar

## Facets
- artifact type: service
- maturity: maintenance
- function: llm-inference, machine-learning, monitoring, container-orchestration
- domain: machine-learning, deep-learning, self-hosted
- platform: python, cloud
- tags: model-serving, torchserve, inference-server, pytorch, mlops, model-archiver, limited-maintenance, docker, kubernetes, linux, gpu

## Member repositories
- pytorch/serve (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:46.198715+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:21:34.925293+00:00, confidence not recorded.
  - readme: https://github.com/pytorch/serve (fetched 2026-08-28T04:08:46.198715+00:00, sha 5385cb38499b)
  - homepage: https://pytorch.org/serve/ (fetched 2026-08-29T09:10:35.242152+00:00, sha 648157cd4055)
  - site_page: https://pytorch.org/docs/stable/index.html (fetched 2026-08-29T09:10:35.252186+00:00, sha a9a24195924a)
  - site_page: https://pytorch.org/ecosystem (fetched 2026-08-29T09:10:35.244880+00:00, sha b1e60b6a50af)
  - site_page: https://discuss.pytorch.org/ (fetched 2026-08-29T09:10:35.246647+00:00, sha 5a4b54973661)
  - site_page: https://pytorch.org/edge (fetched 2026-08-29T09:10:35.248479+00:00, sha f163d0fd8a9e)
  - site_page: https://pytorch.org/executorch/stable/index.html (fetched 2026-08-29T09:10:35.250234+00:00, sha ed8dcff0982a)
  - site_page: https://pytorch.org/pytorch-domains (fetched 2026-08-29T09:10:35.253778+00:00, sha 7c14ee568014)
  - site_page: https://pytorch.org/foundation (fetched 2026-08-29T09:10:35.256737+00:00, sha 826b32be4d07)
- Data as of 2026-08-30T08:39:29.467469+00:00.