TensorFlow Serving

license

Apache-2.0(OSI)

source

public(the published repository is the engine)

core-gated

ungated(no enterprise path in the repository root and no paid build of the engine in the README)

The LICENSE body is the Apache-2.0 text, read in full rather than taken from the API's label, and carries no appended condition. The repository is public and unarchived and builds the engine itself. Its root tree carries no enterprise, ee or commercial directory and its README describes no licence-gated build, so the core reads as ungated. Confidence is medium because that is a repository-and-README read rather than a pricing-page read.

• https://raw.githubusercontent.com/tensorflow/serving/master/LICENSE recorded 2026-08-31

  LICENSE body, read in full: the stock Apache-2.0 text, appendix filled in for The TensorFlow Serving Authors, with no appended condition.

• https://api.github.com/repos/tensorflow/serving recorded 2026-08-31

  Repo metadata for tensorflow/serving - archived false, 6,360 stars.

• https://raw.githubusercontent.com/tensorflow/serving/master/README.md recorded 2026-08-31

  README, read for a paid tier, an enterprise edition or a licence-gated build beside the published source, and for whether any component is stated to be private.

6,360 GitHub stars, which lands in the 1K-10K band of the stars scale, level 2, and a stars-based signal caps at 3 because a star is not a use. Two larger figures exist and neither is this product's. The `tensorflow-serving-api` package draws 4,627,937 downloads in the trailing 30 days, but its own summary is "TensorFlow Serving Python API" - it is the client library for a running server, so its downloads measure client use rather than deployments of the serving system, and it is not declared as an artifact for that reason. The `tensorflow/serving` container image the README's quickstart pulls reports 80,659,067 pulls, which is the product itself, but the count is cumulative since publication while this scale is monthly, so it corroborates without banding. No deployment figure is published, leaving stars as the only honest instrument.

• https://api.github.com/repos/tensorflow/serving recorded 2026-08-31

  Repo metadata - stargazers_count = 6,360 - for tensorflow/serving.

• https://pypi.org/pypi/tensorflow-serving-api/json recorded 2026-08-31

  PyPI metadata for the client library, cited to record why it is not banded: summary "TensorFlow Serving Python API.", author Google Inc., homepage tensorflow.org/serving.

• https://hub.docker.com/v2/repositories/tensorflow/serving/ recorded 2026-08-31

  Docker Hub metadata: pull_count 80,659,067, cumulative since publication. Corroborating only.

Two bands below the vllm anchor. It is a general model server designed before the LLM serving techniques the anchor is banded on - paged attention, continuous batching, tensor parallelism - and documents none of them.

• https://raw.githubusercontent.com/tensorflow/serving/master/README.md recorded 2026-08-31

  README, read for the engine's feature set: Model lifecycle management with a batching scheduler and gRPC/HTTP inference endpoints; extensible beyond TensorFlow models.