PyTorch Foundation

The detailed audience, self-hosting, pricing, and restrictions review has not yet been completed. Consult the sourced product records below and the publisher’s current terms. Missing detail means unknown, not unavailable.

Website

pytorch.org (external site: pytorch.org)

Ownership (separate from openness)

Foundation-hosted

Recorded classification; fact-level ownership verification pending.

Legal form

Directed fund of The Linux Foundation2

Headquarters

Not documented

Founded

20223

Hosts open-source projects. PyTorch itself is published under a BSD-style license in a public repository.15

No product records yet.

• Parent organization: The Linux Foundation (Hosted project)

  Fact-level relationship verification pending.

Units and subsidiaries have their own records and are counted separately from their parent.

• The founding governing board members were AMD, Amazon Web Services, Google Cloud, Meta, Microsoft Azure, and NVIDIA.3
• A Governing Board oversees the foundation under its charter. Technical governance of the PyTorch project is defined separately by its maintainers and is open to individuals, not companies.14
• The PyTorch code base is established as PyTorch a Series of LF Projects, LLC, and foundation members must also be Linux Foundation members.2

1. Oct 18, 2026 · Release

   vLLM 0.32.0 released with a structured decisions endpoint and wider low-precision KV cache support

   On October 18, 2026 the vLLM project published version 0.32.0 on GitHub. The release notes highlight a new POST /v1/systemone endpoint that answers choice, noul and score questions over a structured state, and a new hf tool and reasoning parser that reads a checkpoint's response_template so models that ship it need no model-specific parser. The notes also describe NVFP4 KV cache support on more GPU architectures and new model support, including EmbeddingGemma 2, MiniCPM-V 4.7 and Nemotron 3.5 ASR. Under security, they list fixes for three advisories and say CORS no longer allows every origin by default; breaking changes include dropping Python 3.10 and removing cascade attention. USASI has not tested any of these changes.

   • PyTorch Foundation
   • vLLM

2. Oct 5, 2026 · Release

   vLLM 0.31.0 released with a preload command for faster engine restarts

   On October 5, 2026 the vLLM project published version 0.31.0 on GitHub. The release notes highlight a new vllm preload command that launches a weight-cache daemon to keep post-quantized weights in GPU memory across engine restarts, and experimental engine snapshots (vllm snapshot create/restore) that use CRIU to restore a fully initialized TP1 engine. Under security, the notes say per-request mm_processor_kwargs and media_io_kwargs are now rejected unless --trust-request-mm-kwargs is set. The release also lists breaking changes, including the removal of tokenizer_mode="slow", and performance work for DeepSeek-V4.1-Flash. USASI has not tested any of these changes.

   • PyTorch Foundation
   • vLLM

3. Sep 2, 2026 · Release

   PyTorch 2.14.0 released with torch.switch, NVGEMM in Inductor, and a preview NCCL backend rewrite

   PyTorch 2.14.0 was published on GitHub on September 2, 2026. The release notes highlight NVGEMM, which brings CuTeDSL-generated CUTLASS kernels to Inductor, and torch.switch, which generalizes torch.cond to multi-way branching. Other highlights include declarative dynamic shapes via @dynamic_spec, experimental torch.compile support for complex-valued tensors, a preview of a rewritten NCCL backend ported from torchcomms, and native linear algebra on Apple Silicon. The notes also say Inductor now targets Rubin (sm_107). USASI has not tested this release.

   • PyTorch Foundation
   • PyTorch

4. Jul 8, 2026 · Release

   PyTorch 2.13.0 released with FlexAttention on Apple Silicon and the torchcomms backend

   PyTorch 2.13.0 was published on GitHub on July 8, 2026. The release notes highlight FlexAttention support on Apple Silicon (MPS) and a deterministic FlexAttention backward path on CUDA, a prototype CuTeDSL "Native DSL" backend for Inductor alongside Triton, and nn.LinearCrossEntropyLoss, which combines the final prediction and loss computation to reduce peak GPU memory. They also introduce torchcomms, a new communications backend for PyTorch Distributed, and Python 3.15 wheels on Linux, while dropping free-threaded CPython 3.13t binaries and Bazel build support. USASI has not tested this release.

   • PyTorch Foundation
   • PyTorch

• Legal name: not documented in the sources reviewed for this record.
• Headquarters: not documented in the sources reviewed.
• Official careers page: not verified by this catalog.
• Products: none documented in this record.

Have a primary source? How to report a correction.

• HubBuilding software with AI
• HubOpen-source foundations and stewards
• ExplainerHow the American AI ecosystem fits together

Eligibility policy

1. 1.
   PyTorch Foundation (external site: pytorch.org)

   PyTorch Foundation · Official page · accessed Sep 29, 2026

2. 2.
   PyTorch Foundation Participation Agreement and Charter (external site: cdn.platform.linuxfoundation.org)

   The Linux Foundation · Official page · accessed Sep 29, 2026

3. 3.
   Meta Transitions PyTorch to the Linux Foundation, Further Accelerating AI/ML Open Source Collaboration (external site: linuxfoundation.org)

   Linux Foundation · Announcement · published Sep 12, 2022 · accessed Sep 29, 2026

4. 4.
   PyTorch Governance | Mechanics (external site: docs.pytorch.org)

   PyTorch Foundation · Documentation · accessed Sep 29, 2026

5. 5.
   pytorch/pytorch README (external site: github.com)

   PyTorch (GitHub) · Repository · accessed Sep 29, 2026

6. 6.
   About the Linux Foundation (external site: linuxfoundation.org)

   Linux Foundation · Official page · accessed Sep 29, 2026

7. 7.
   Privacy Policy | Linux Foundation (external site: linuxfoundation.org)

   Linux Foundation · Official page · accessed Sep 29, 2026