AutoRound

license

Apache-2.0(OSI)

source

public

core features withheld

no

Apache-2.0 license body confirmed. The repository is public and unarchived and builds the whole product, and the README describes no paid tier, enterprise edition or license-gated build beside it, so source is public and the core ungated.

• https://api.github.com/repos/intel/auto-round recorded 2026-09-02

  Repo metadata for auto-round's canonical repository - private false, archived false, fork false.

• https://github.com/intel/auto-round/blob/main/LICENSE recorded 2026-10-07

  LICENSE body carries the verbatim Apache License 2.0 text under an Intel copyright header.

• https://github.com/intel/auto-round/blob/main/README.md recorded 2026-10-07

  README describes AutoRound as "an easy-to-use post-training quantization toolkit for Large Language Models (LLMs) and Vision-Language Models (VLMs)" that "supports several practical low-bit and adaptive-bit quantization algorithms", with SignRound at its center, deployable "across Intel CPUs and XPUs, Gaudi accelerators, and NVIDIA GPUs"; it names no paid tier, enterprise edition or license-gated build beside the published source.

Adoption is measured from PyPI downloads of the `auto-round` package. The README's `auto-round-nightly` and `auto-round-hpu` variants are not counted, so the figure is a floor on total use.

• https://pypistats.org/api/packages/auto-round/recent recorded 2026-09-02

  last_month downloads = 416,661 for auto-round

AutoRound centers on sign-gradient-descent rounding and carries several other post-training quantization algorithms beside it, applied across a wide range of LLMs, VLMs and Intel and NVIDIA hardware. All of that breadth sits within one optimization family rather than a broader compilation toolchain, which is why it sits two steps below Apache TVM.

• https://github.com/intel/auto-round/blob/main/README.md recorded 2026-10-07

  README describes AutoRound as "an easy-to-use post-training quantization toolkit for Large Language Models (LLMs) and Vision-Language Models (VLMs)" that "supports several practical low-bit and adaptive-bit quantization algorithms", with SignRound at its center, deployable "across Intel CPUs and XPUs, Gaudi accelerators, and NVIDIA GPUs"; it names no paid tier, enterprise edition or license-gated build beside the published source.