main
[doc] Add AOTInductor intermediate debug printer OSS user manual (#163794)
Migrate to new theme (#149331)
[OpenReg][Feat][Docs] Enrich OpenReg device management implementation and add focused documentation (#165897)
Fix link for core maintainers request form (#168089)
Add dynamic shapes doc (#159428)
Support NUMA Binding for Callable Entrypoints (#160163)
Revert "Escaped html tags name and target to appear as strings (#165543)"
Fix typo in FP16 accumulation section (#167703)
Fix broken URLs (#152237)
[ONNX] Filter out torchscript sentences (#158850)
[OpenReg] Migrate Accelerator Document from source/notes into source/accelerator (#161845)
Introduce a new API torch.accelerator.get_memory_info (#156812)
[Docs] Convert to markdown: accelerator.rst, amp.rst, autograd.rst, backends.rst, benchmark_utils.rst (#155762)
[reland] Warn if AccumulateGrad stream does not match producer node stream (#166136)
Add split-K control to cuBLAS reduced-precision settings (#164766)
[Fix] fix gramma error in PyTorch docs (#166158)
Fix typos in complex numbers docs (#166671)
[Docs] Fix indentations in cond.md (#156147)
[BE] adding documentation (#167334)
[Docs] Convert to markdown cond.rst, config_mod.rst (#155653)
xpu: support sycl with torch.utils.cpp_extension APIs (#132945)
[3/n] Remove references to TorchScript in PyTorch docs (#158315)
[BE] document distributed apis (#165194)
Fix typos under docs directory (#88033)
[BE] Adding aliases for CUDA and XPU API documentation (#162984)
[CUDA] Add experimental green context support for SM carveout (#159104)
[ROCm][tunableop] Improvements to tunableop Numerical Check (#163079)
Add doc page for environment variables that effect PyTorch Runtime (#119087)
Remove orphan from cuDNN persistent note (#65160)
Fix #155016 for Docathon - convert rst to markdown (#155198)
DOC: Convert to markdown: ddp_comm_hooks.rst, debugging_environment_variables.rst, deploy.rst, deterministic.rst, distributed.algorithms.join.rst (#155298)
Add a title to distributed._dist2.md (#159385)
Revert "list_stored_sd_metadata API. (#160610)"
NUMA binding integration with elastic agent and torchrun (#149334)
[FSDP2] provide public API to share cuda streams across roots (#165024)
distributed/debug: add an HTTP server for debugging running jobs (#167395)
Fix #155018 (convert distributed rst to markdown) (#155528)
[PP] Add DualPipeV schedule (#159591)
Move MaskPartial to placement_types to improve discoverability (#164414)
fix Dtensor doc link (#162494)
Convert to markdown: distributed.tensor.parallel.rst, distributed.tensor.rst, distributions.rst, dlpack.rst (#155297)
Revert "Revert D21337640: [pytorch][PR] Split up documentation into subpages and clean up some warnings" (#37778)
Update doc (#166024)
Convert to .md: draft_export.rst, export.ir_spec.rst, fft.rst (#155567)
Convert rst files to md (#155369)
Fix #155022 rst to markdown conversion (#155540)
Convert hub.rst to hub.md (#155483)
Add placeholder for the User Guide (#159379)
[4/n] Remove references to TorchScript in PyTorch docs (#158317)
[2/n] Remove references to TorchScript in PyTorch docs (#158306)
[1/n] Remove references to TorchScript in PyTorch docs (#158305)
Convert to markdown: jit_python_reference.rst, jit_unsupported.rst, jit_utils.rst, library.rst (#155404)
Add utility to get computed kernel in torch.library (#158393)
[Docs] Convert to markdown to fix 155025 (#155789)
adding quantization.rst file for quantization feature (#27559)
DOC: Convert to markdown: mobile_optimizer.rst, model_zoo.rst, module_tracker.rst, monitor.rst, mps_environment_variables.rst (#155702)
Fix/issue #155027 (#155252)
[MTIAGraph][Pytorch][3/n] Implement mtia_graph python wrapper in pytorch (#166964)
Convert to markdown: named_tensor.rst, nested.rst, nn.attention.bias.rst, nn.attention.experimental.rst, nn.attention.flex_attention.rst #155028 (#155696)
[nativert] aoti (#162353)
[BE] Use .md instead of .rst for nn.aliases doc (#158666)
[BC Breaking] Remove flex + njt code paths (#161734)
Add FA4 to sdpa (#167348)
varlen api (#164502)
Add scaled_grouped_mm_v2 and python API (#165154)
update nn.init doc to reflect the no_grad (#80882)
[BE] More torch.nn docs coverage test (except for torch.nn.parallel) (#158654)
[ONNX] Expose the testing module (#162495)
[ONNX] Remove enable_fake_mode and exporter_legacy (#161222)
[ONNX] Implement Attention-23 (#156431)
[ONNX] Refactor torchscript based exporter (#161323)
Document the rest of the specific optimizer module APIs (#158669)
[muon] Introduce Muon optimizer to PyTorch (#160213)
Convert rst to markdown - profiler.rst #155031 (#155559)
[BE] document some quantization public apis (#165160)
Convert to markdown: quantization-accuracy-debugging.rst, quantization-backend-configuration.rst, quantization-support.rst, random.rst (#155520)
RPC tutorial audit (#157938)
Convert rst to md: rpc.rst, signal.rst, size.rst, special.rst (#155430)
[BE] fix typos in docs/ (#156080)
Super tiny fix typo (#151212)
Add doc for Symmetric Memory (#166148)
revamp dtype documentation for 2025 (#156087)
[docs] fix numpy docs reference (#147697)
Cleanup all module references in doc (#73983)
Convert to markdown: testing.rst, threading_environment_variables.rst, torch_cuda_memory.rst, torch_environment_variables.rst, torch_nccl_environment_variables.rst (#155523)
Revert "Remove torch.serialization entries from the doc ignore list (#160224)"
dynamo configs to torch.compiler (#163517)
[AOTI] Update AOTInductor tutorial (#163808)
[doc] AOTI debugging guide (#160430)
Converting .rst files to .md files (#155377)
[dynamo] Make global state guards and torch function stack guards droppable. (#167674)
Add AOTDispatcher config to set backward autocast behavior (#156356)
[Graph Partition] add graph partition doc (#159450)
DOC: Convert to markdown: torch.compiler_best_practices_for_backends.rst, torch.compiler_cudagraph_trees.rst, torch.compiler_custom_backends.rst, torch.compiler_dynamic_shapes.rst, torch.compiler_dynamo_deepdive.rst (#155137)
[ez] fix small doc error (#164915)
Dynamo Deep Dive Documentation Fix (#158860)
convert: rst to myst pr 1/2 (#155840)
convert: rst to myst pr2/2 (#155911)
Convert compiler rst files to markdown (#155335)
Add to inductor provenance tracking doc (#162975)
[export] Update docs (#157750)
[Docs] Update PT2 Profiler Torch-Compiled Region Image (#158066)
[Docs] Convert to markdown: torch.compiler_transformations.rst, torch.compiler.config.rst (#155347)
Add torch compile force disable caches alias (#158072)
DOC: Convert to markdown: torch.overrides.rst, type_info.rst, utils.rst, xpu.rst (#155088)
Fixes broken memory_viz link in CUDA memory docs (#161426)
finfo eps doc fix (#160502)
Rename to _debug_mode.py to make it private (#163534)
Introduce a new API torch.xpu.get_per_process_memory_fraction (#165511)
