Triton is the default GPU code generator inside torch.compile/TorchInductor (since PyTorch 2.0), with NVIDIA and AMD backends in tree and an Intel GPU backend maintained out of tree
Default since PyTorch 2.0 (Mar 2023)observed
| Value kind | observed — Reported measurements, counts, and specifications keep the precision and scope stated by their source; an exact specification is not treated as a range. |
|---|---|
| As of | 2023-03 |
| Source | PyTorch 2.0 release announcement (PyTorch Foundation, 2023-03-15) · Release blog, Summary and '[Beta] torch.compile': TorchInductor underpins torch.compile and, for NVIDIA and AMD GPUs, uses OpenAI Triton to generate kernels. Default backend: PyTorch 2.0 Get Started page ('By default, TorchInductor is used'). Current status incl. Intel GPU backend: torch.compiler docs, https://docs.pytorch.org/docs/2.9/torch.compiler.html |
| Review | checking…review by 2027-06-18 · slow cadence |
| Recorded changes | last 2026-10-19 · 2 revisions tracked |
| Claim id | triton-is-the-default-gpu-code-generator-inside |
Where the guide uses it
← Full numbers register — every date-stamped figure in the guide, with revision history.