518c2b1

[10/N] Use Python 3.10 typing (#169229)

cyyever Dec 2, 2025

1902edd

[precompile] support serde for torch.nn.attention.SDPBackend (#168988)

bobrenjc93 Dec 1, 2025

587d63a

[precompile] generate nonce key if enable_aot_compile is enabled (#16…

bobrenjc93 Dec 1, 2025

ce5e7e3

[CI][CUDA][Distributed] Update NCCL to 2.28.9 for CUDA13 (#168091)

nWEIdia Dec 2, 2025

2d1f78f

[DebugMode] Fix hash for 0 ele tensor; Add more tests (#169027)

yushangdi Dec 2, 2025

503b264

[AOTI] Fix a GPU memory leak caused by reference circle (#168063)

desertfire Dec 1, 2025

3cd98b4

[Inductor] support masked vectorization for the tail_loop for integer…

jiayisunx Dec 1, 2025

d9cb8a7

[AOTI] Set device info for subgraphs (#169001)

desertfire Dec 1, 2025

1e526fb

Parity of rng offset compute and ranks subset support for Local Tenso…

dzmitry-huba Dec 1, 2025

c04e2c6

[dynamo][dicts] Decentralize and Improve key hash implementation for …

anijain2305 Dec 1, 2025

d766976

[Inductor] [Triton] Capture Timeout errors without crashing the job (…

njriasan Dec 2, 2025

bb30341

Avoid std::tie and returning value constructions in qconv_unpack.cpp …

cyyever Dec 2, 2025

fb5be22

Refactor: Remove unnecessary ConstantVariable wrapping in raise_obser…

nandan2003 Dec 2, 2025

166efda

[MTIAGraph][Pytorch] Add the graph_pool_handle api (#169283)

andyanwang Dec 2, 2025

62d3ccd

[audio hash update] update the pinned audio hash (#169198)

pytorchupdatebot Dec 2, 2025

7d1bbaf

Add sum support for qlinear_binary templated implementation (#163249)

CaoE Dec 2, 2025

66004b9

[dynamo] Fix test state leakage for test_modes.py (#168928)

jansel Dec 2, 2025

b9c8f3a

[dynamo] Fix local test failures for logging (#168927)

jansel Dec 2, 2025

4cfb47f

[CP] Refactor CP sharding rules into separate module and only registe…

fegin Dec 1, 2025

f2d6a75

bump timm pin (#169227)

BoyuanFeng Dec 2, 2025

b555c39

[pallas backend] Add multi-output support to Pallas backend (#169323)

oulgen Dec 1, 2025

afdff7f

[pallas backend] support bitcast (#169324)

oulgen Dec 1, 2025

a951a9c

[inductor] Increase tolerance for test_emulate_precision_casts_mean_r…

jansel Dec 2, 2025

556375b

[inductor] Increase tolerance for test_conv3d_binary_broadcast_shapes…

jansel Dec 2, 2025

1e34fb2

Revert "Refactor: Remove unnecessary ConstantVariable wrapping in rai…

pytorchmergebot Dec 2, 2025

4917316

[Dynamo][Guards]Fix TLParse CPP guard message with sorting get_leaf_g…

fxdawnn Dec 2, 2025

5778f6f

[strobelight][gpusnoop] compress aoti stack (#169291)

sonntex Dec 2, 2025

892640e

Hide all symbols (except stable/headeronly/shim) if TORCH_STABLE_ONLY…

mikaylagawarecki Dec 1, 2025

70d797a

[dist] add reduce_scatter_out (#168260)

IvanKobzarev Dec 2, 2025

285779b

[Accelerator] Add Accelerator Capabilities API (#165631)

can-gaa-hou Dec 2, 2025

174272c

[precompile] Include inductor-generated Dynamo guards in AOT (#169239)

zhxchen17 Dec 2, 2025

9ff4a2e

Support AC in default partitioner when functionalization is enabled (…

soulitzer Dec 1, 2025

44ac693

Add DCE pass to remove unused intermediates (#169131)

tugsbayasgalan Dec 1, 2025

b4cc132

[BE][Typing][Dynamo] Type torch/_dynamo/variables/nn_module.py (#167342)

Lucaskabela Dec 2, 2025

1aa13e1

[dynamo] Fix more local test failures (#169228)

jansel Dec 2, 2025

c6ae757

[Submodule] Update to cutlass 4.3 (#168308)

drisspg Dec 1, 2025

93d0d68

Triton 3.6 pin update (#168096)

atalman Dec 2, 2025

7a41b66

[Inductor] handle GroupedSchedulerNode in combo kernel fusion (#168109)

karthickai Dec 2, 2025

1cee47d

REmove AC specific flag and reunify with HOPs (#169132)

tugsbayasgalan Dec 2, 2025

7741edd

[MPS] Fix dlpack exports/imports for sliced tensors (#169272)

malfet Dec 1, 2025

4e0061c

[ez] add return type to _unpickle_sdp_backend (#169383)

bobrenjc93 Dec 2, 2025

7007646

Revert "Remove unnecessary uses of thrust::tuple (#168936)"

pytorchmergebot Dec 2, 2025

6f7dcf5

Add shape logging to autotuning to debug timeout issues (#169062)

njriasan Dec 2, 2025

066997f

[CUDA][FlexAttention] Use `sm8x` configs for `sm12x` for backward (#1…

eqy Dec 2, 2025

d973dc6

[ez] Remove maybe unused var in common_utils.py (#166455)

clee2000 Dec 2, 2025

2c87367

dist/debug: add TCPStore debug page (#169095)

d4l3k Dec 2, 2025

d998c03

dist/debug: use aiohttp to scale to 100k workers (#169096)

d4l3k Dec 2, 2025

082e96b

distributed/debug: add fr trace analysis (#169144)

d4l3k Dec 2, 2025

6f2783a

[MTIA] Support bfloat16 in half_to_float (#168938)

patrick-toulme Dec 2, 2025

8d9dd96

ensure that regional inductor uses a boxed calling convention (#168277)

bdhirsh Dec 1, 2025

b7d6068

fix for using cross-pg overlap (#169384)

eellison Dec 2, 2025

d8fd5c6

[MPS] Modify eps for gradcheck for float32 dtypes (#168902)

Isalia20 Dec 2, 2025

fec710b

Triton 3.5+fb: Drop the legacy support for autoWS (#169089)

agron911 Dec 2, 2025

15da210

[torch.export] Copy common custom metadata of children node to parent…

malaybag Dec 2, 2025

c0660bc

[CI] Run inductor-unittests if workflow file is modified (#169398)

malfet Dec 2, 2025

6ec30b4

[xplat][caffe2] Fix -Wswitch-default issues (#169022)

NSProgrammer Dec 2, 2025

81af382

[GB300] [Triton] Disable experimental Triton API inside PyTorch when …

njriasan Dec 2, 2025

45d14e2

[ROCm] Enable ZerO Optimizer UTs (#169077)

pragupta Dec 2, 2025

8ef0c0b

dist/debug: support py-spy (native+subprocess) stacks (#169147)

d4l3k Dec 2, 2025

31fc127

[dynamo] add torch._dynamo.set_recursion_limit to fix 3.12/3.13 Recur…

williamwen42 Dec 1, 2025

7f55ba1

[dynamo, guards] add guard builder microbenchmark (#169087)

williamwen42 Dec 1, 2025

ef8ecc1

[11/N] Use Python 3.10 typing (#169335)

cyyever Dec 3, 2025

0c281dd

Revert "[MPS] Fix dlpack exports/imports for sliced tensors (#169272)"

pytorchmergebot Dec 3, 2025

d1c9f03

[4/N] Use context managers (#169169)

cyyever Dec 3, 2025

d19f1e8

Remove unnecessary uses of thrust::tuple (#168936)

cyyever Dec 3, 2025

6c261c6

Revert "[Accelerator] Add Accelerator Capabilities API (#165631)"

pytorchmergebot Dec 3, 2025

0b80a4c

[dynamo, ci] Update 3.13 pull CI tests to Python 3.14 (#169032)

guilhermeleobas Dec 2, 2025

ef019d1

Silent XNNPACK GCC14 warnings (#166873)

fadara01 Dec 2, 2025

7a1e316

Fix missing ConstantPooling header in passes.cpp (#169420)

huydhn Dec 3, 2025

c178ed4

Revert "[Dynamo][Guards]Fix TLParse CPP guard message with sorting ge…

pytorchmergebot Dec 3, 2025

59abd50

[reland][Full Inductor][Pytorch] Prevent decomposition to support fal…

andyanwang Dec 3, 2025

3418bd2

Revert " [10/N] Use Python 3.10 typing (#169229)"

pytorchmergebot Dec 3, 2025

135f375

[xpu][feature] [3/3] Register the `scaled_mm` and `scaled_mm_v2` for …

Stonepia Dec 3, 2025

5fafc13

[MPS] Fix dlpack exports/imports for sliced tensors (#169272)

malfet Dec 1, 2025

2bec68e

[dynamo][hops] Remove restore_side_effect from the auto flattening co…

anijain2305 Dec 3, 2025

bac403c

[Dynamo] Adding `strict` as a kwarg for `map()` in Python3.14 (#167828)

can-gaa-hou Dec 3, 2025

74fe26a

Fixes complex datatype handling in ddp (#166863)

arkadip-maitra Dec 3, 2025

0bbbdf1

[Inductor] Fix Diode / exhaustive autotune crash on AMD (#169225)

JChunX Dec 3, 2025

5bf1cdf

[DTensor] compute shape and offset for arbitrary _StridedShard (#168146)

weifengpy Dec 2, 2025

47b28dd

Eliminate GPU to CPU sync in linalg.eig for CUDA tensors (#168283)

razaaliraza Dec 3, 2025

5f21d27

[xpu][feature] Upgrade XPU OneDNN to v3.10.2 (#169443)

LuFinch Dec 3, 2025

3d35fd2

Fix device determination logic in Conditional (#169199)

larryliu0820 Dec 3, 2025

6ff8311

[xpu] Enable TransformerEncoderLayer Fast Path for XPU Device (#168234)

yucai-intel Dec 3, 2025

9b3e34d

[MPS][BugFix] Fix MaxPool2d/MaxPool3d output size validation (Issue #…

lingebeng Dec 3, 2025

6ceb4a3

Remove `LocalGeneratorObjectVariable::_get_inline_tracer()` (#169306)

guilhermeleobas Dec 3, 2025

78adb3b

[generator] Close all open generators in compile_subgraph (#157149)

guilhermeleobas Dec 3, 2025

9f0df56

Restore LD_LIBRARY_PATH after binary wheel checks (#169444)

atalman Dec 3, 2025

8d56e98

Smoke tests for binary symbol hiding (#169307)

mikaylagawarecki Dec 1, 2025

597930f

Correctly set max_numwarps in coordinate_descent_tuner (#159146)

jataylo Dec 3, 2025

ecbcc3f

[Fix] Add safeguard for unsafe index (#169234)

lingebeng Dec 3, 2025

ee87bbe

Support AC rematerializing in forward+loss+bwd graph (#168082)

tugsbayasgalan Dec 3, 2025

7ba4680

Fix deprecation warning (#169421)

drisspg Dec 2, 2025

fe0e65a

Enabling WOQ fusion path on ACL (#165643)

gassan-arm Dec 3, 2025

e3f24fd

Fix test_matmul_mv_cpu_float32 and Enable test_linalg for AArch64 (#1…

robert-hardwick Dec 2, 2025

18f3ca0

[ROCm] Enable StaticCudaLauncher for ROCm (#166492)

chinmaydk99 Dec 3, 2025

7348cb3

[Submodule] cutlass bump to minor bug fix (#169433)

drisspg Dec 2, 2025

4ae6846

Add inductor core tests for Python 3.11/3.12 (#167542)

guilhermeleobas Dec 2, 2025

0907694

Run dynamo/inductor core tests on Python 3.13 (#168293)

guilhermeleobas Dec 2, 2025

07dcc0b

[ROCm][CI] Add docker caching for MI250 runners (#169300)

jithunnair-amd Dec 3, 2025

f7e1bd8

Pass GPU_FLAGS inside container during the test (#169397)

malfet Dec 3, 2025

1d21b4d

Add more local test coverage for DTensor based tests (#169396)

dzmitry-huba Dec 2, 2025

8732949

Add shims for setCurrentCUDAStream, getStreamFromPool, and synchroniz…

pearu Dec 3, 2025

d038b01

The Nested Pool (#168382)

eee4017 Dec 3, 2025

52ac0f0

[CI] Run both CPU and CUDA Pallas tests

malfet Dec 3, 2025

2887faa

[inductor_on_demand] Refactor - scoop regions and then compile region…

anijain2305 Dec 3, 2025

abfa1a6

[aot_compile] Pass additional globals to callable (#169070)

angelayi Dec 3, 2025

8c73bbb

[MPS] Migrate `clamp.Tensor_out` to Metal (#169407)

kurtamohler Dec 3, 2025

39d07db

Use stable topological sort in fuse_by_partitions (#167397)

SherlockNoMad Dec 3, 2025

195f92e

[pallas backend] fix breakages (#169406)

oulgen Dec 2, 2025

dfbd371

Enable custom collective op autotuning (#167294)

tianrengao Dec 3, 2025

d40f495

[hoo] Invoke subgraph + effects Inductor support (#167364)

angelayi Dec 3, 2025

201e2c4

Fix update slow test workflow after #168334 (#169495)

atalman Dec 3, 2025

bc6a486

[xpu][feature][2/N]Enable SDPA XPU FlashAttention backend with SYCL-T…

LuFinch Dec 3, 2025

34a9860

[MPS] Add `linalg.lu_solve` and `linalg.lu` (#167569)

kurtamohler Dec 3, 2025

89e3bbc

[Accelerator] Add Accelerator Capabilities API (#165631)

can-gaa-hou Dec 3, 2025

e7d24d3

[dynamo, 3.14] Don't use _PyObject_GC_TRACK which is internal and cal…

DinoV Dec 3, 2025

61be54a

update the MHA to enable cudnn-frontend 1.12.1 (#169086)

cicichen01 Dec 3, 2025

fdf863d

Revert "Triton 3.6 pin update (#168096)"

pytorchmergebot Dec 3, 2025

296e67c

[effect] Remove special handling for profiler op (#168389)

angelayi Dec 3, 2025

6e404e9

Compiler User Guide Landing Pages (#165635)

sekyondaMeta Dec 3, 2025

eb5c636

Generalize GraphView in manual bucketing (#169426)

yushangdi Dec 3, 2025

7eb6259

[dynamo][benchmarks] add option to force amp to use bfloat16 instead …

xmfan Dec 3, 2025

f9bd6c5

Support module.to in strict export (#167555)

tugsbayasgalan Dec 2, 2025

ec2c71f

[CI][Docker] Add triton to pallas build (#169494)

oulgen Dec 4, 2025

3051687

Revert "Enable custom collective op autotuning (#167294)"

pytorchmergebot Dec 4, 2025

65c4620

Revert "[effect] Remove special handling for profiler op (#168389)"

pytorchmergebot Dec 4, 2025

320de0c

Add public documentation for stable_topological_sort (#169498)

SherlockNoMad Dec 3, 2025

6c8b6a0

localtensor pointwise ops test (#169500)

dolpm Dec 4, 2025

2df6058

[opaque_obj] Remove free registration (#167739)

angelayi Dec 3, 2025

7cbc2d0

[opaque_obj] Remove inital opaque obj (#167740)

angelayi Dec 3, 2025

d16447d

[opaque obj] Set type of FakeScriptObj to be OpaqueObj type (#167741)

angelayi Dec 3, 2025

d78f52b

[opaque obj] Improve error msg for intermediate opaques (#167742)

angelayi Dec 3, 2025

bea4912

[Dynamo][Guard]Add the user-friendly TYPE_MATCH for type (#169025)

fxdawnn Dec 3, 2025

b3a7edb

Revert "Add public documentation for stable_topological_sort (#169498)"

pytorchmergebot Dec 4, 2025

b2b6b03

[export] Make RNNs exportable on GPUs (#163245)

yiming0416 Dec 4, 2025

2ac3ef8

Revert "[generator] Close all open generators in compile_subgraph (#1…

pytorchmergebot Dec 4, 2025

e115f9f

Revert "Support module.to in strict export (#167555)"

pytorchmergebot Dec 4, 2025

85a3159

Enable custom collective op autotuning (#167294)

tianrengao Dec 4, 2025

2e0c2e1

[xpu][feature] [1/2] Introduce XPUPluggableAllocator in cpp part (#16…

guangyey Dec 3, 2025

7b7af39

[MPS] Migrate `lu_unpack` to Metal and fix `lu_solve` backward (#168120)

kurtamohler Dec 3, 2025

02d8bd6

[pallas backend] Add special function support to Pallas backend (#169…

oulgen Dec 4, 2025

d900f5e

Update slow tests (#167967)

pytorchupdatebot Dec 4, 2025

49a04d2

[effect] Remove special handling for profiler op (#168389)

angelayi Dec 3, 2025

bc39b2b

Replace `msg` by `args` in `raise_observed_exception` (#169343)

guilhermeleobas Dec 2, 2025

6f53fef

Revert "Avoid std::tie and returning value constructions in qconv_unp…

pytorchmergebot Dec 4, 2025

a15066c

Fix torch.fx for the newer "|" union syntax (#169453)

cyyever Dec 4, 2025

cc0853a

Revert "Replace `msg` by `args` in `raise_observed_exception` (#169343)"

pytorchmergebot Dec 4, 2025

dd18a75

[xpu][feature] [2/2] Introduce XPUPluggableAllocator in frontend part…

guangyey Dec 3, 2025

597df3a

[CI] Use py3.10 in upload-utilization-stats (#169501)

desertfire Dec 3, 2025

b1decff

Revert "Fix torch.fx for the newer "|" union syntax (#169453)"

pytorchmergebot Dec 4, 2025

5191b2f

[AOTI] Fix a small buffer mutation issue (#169347)

desertfire Dec 3, 2025

1c87554

[CI] Add CUDA 13.0 inductor CI benchmarks (#165029)

tinglvv Dec 4, 2025

d54ff42

[dynamo, guards] apply functools.cached_property to Source.name (#168…

williamwen42 Dec 3, 2025

bc43d5b

[dynamo, guards] cache GuardBuilder.get() on sources (#168203)

williamwen42 Dec 3, 2025

ea7035f

[dynamo, guards] cache Source.guard_source (#168386)

williamwen42 Dec 3, 2025

c55b1e8

[dynamo, guards] cache Source hashing (#168886)

williamwen42 Dec 3, 2025

1afe283

[dynamo] Support Sequence-like placement user defined objects from_lo…

anijain2305 Dec 4, 2025

b6b6c80

Revert "[opaque obj] Improve error msg for intermediate opaques (#167…

pytorchmergebot Dec 4, 2025

ba14125

Revert "[dynamo][dicts] Decentralize and Improve key hash implementat…

pytorchmergebot Dec 4, 2025

2353a0f

[dynamo] support tracing DeviceMesh._flatten (#169530)

williamwen42 Dec 4, 2025

b6b6d91

Revert "[Dynamo][Guard]Add the user-friendly TYPE_MATCH for type (#16…

pytorchmergebot Dec 4, 2025

43b9471

Fix torch.fx for the newer "|" union syntax (#169453)

cyyever Dec 4, 2025

76aeb8c

[DebugMode] default values for outputs, stack trace (#169499)

pianpwk Dec 4, 2025

4fefb8e

[OpenReg][CI][Refactor] Refactor OpenReg CI for better independency a…

KarhouTam Dec 4, 2025

ae3a239

Revert "[export] Make RNNs exportable on GPUs (#163245)"

pytorchmergebot Dec 4, 2025

5634469

ci: Minimize dependency on build_environment for setup-nvidia action …

seemethere Dec 3, 2025

35b7a9a

ci: Remove errant nogpu conditional (#169431)

seemethere Dec 3, 2025

ffd9b0f

Resolve collective autotuning test failure on arm (#168919)

tianrengao Dec 4, 2025

eabb7ad

[Inductor XPU GEMM] Step 1/N: Refactor cutlass configuration. (#160174)

etaf Dec 2, 2025

7716da9

[10/N] Use Python 3.10 typing (#169229)

cyyever Dec 4, 2025

c0cb6e7

[DTensor] ExplicitRedistributionContext warning mode (#169452)

wconstab Dec 4, 2025

685ba6b

add back legalize_graph for BC reason (#169541)

SherlockNoMad Dec 4, 2025

a2b5dfb

Restore ability to use global stableivalue from without template (#16…

mikaylagawarecki Dec 3, 2025

f53e14c

[ci] Update typing-extensions for python 3.14 (#169515)

rtimpe Dec 4, 2025

2da3baf

[Inductor] ReLU/GELU(Addmm) fusions (#168157)

nikitaved Dec 4, 2025

da2e3c4

Strengthen the implementation of OpenReg Stream (#166115)

zhudada0120 Dec 4, 2025

a36e1d3

Triton 3.6 pin update (#168096)

atalman Dec 4, 2025

c19ecc7

fix typo: ommunication -> communication (#169557)

crcrpar Dec 4, 2025

005f9fb

Switch off of deprecaated API (#169517)

drisspg Dec 3, 2025

8b4f89e

[CI] Simplify CPython installation (#169510)

malfet Dec 4, 2025

324a828

[c10d][Sym mem] Add set_signal_pad_size API for SymmetricMemory (#169…

yang-yu-hang Dec 4, 2025

e770c95

[inductor] require shape in TritonCSEVariable (#162275)

isuruf Dec 4, 2025

d4ad7a1

[BE] Don't make typos in build-environment between build and test (#1…

malfet Dec 4, 2025

31987d0

[inductor] skip the r2r determ test in fbcode (#169074)

shunting314 Nov 25, 2025

87b9744

Clarify checkpointing docs (#169007)

jder Dec 4, 2025

ada0665

Add C++ wrapper for shim from_blob in stable/csrc/ops.h (#168380)

mikaylagawarecki Dec 4, 2025

d3b4475

Add STD_CUDA_{KERNEL_LAUNCH_}CHECK (#169385)

mikaylagawarecki Dec 4, 2025

6f4baa5

Revert "[ROCm] Enable StaticCudaLauncher for ROCm (#166492)"

pytorchmergebot Dec 4, 2025

e5fd7b7

Add a single GPU variant of modded-nanogpt to torchbench (#169502) (…

xmfan Dec 4, 2025

36f60a9

[dde] use is_contiguous_or_false for function with out= arg (#169305)

ColinPeppler Dec 4, 2025

f575ecb

Revert "Remove unnecessary uses of thrust::tuple (#168936)"

pytorchmergebot Dec 4, 2025

00279ab

[annotate] Change AutogradFunctionApply HOP to use Interpreter to pre…

yushangdi Dec 4, 2025

e64f1ee

expandable_segments + memory pool (#169491)

eee4017 Dec 4, 2025

5aad0e1

[opaque obj] Improve error msg for intermediate opaques (#167742)

angelayi Dec 4, 2025

5c3874b

[ROCm][CI] Enable TD for all ROCm default and distributed config work…

jithunnair-amd Dec 4, 2025

2912ea3

Add huggingface storage reader for MXFP4 quantized GPT-OSS checkpoint…

shuhuayu Dec 4, 2025

31bb133

[dynamo] Refactor isinstance(x, ConstantVariable) to x.is_python_cons…

jansel Dec 4, 2025

45cd3b9

Re-enable torch.compile tests for Python 3.12 and Windows (#169387)

guilhermeleobas Dec 4, 2025

8da5d29

[DLPack] C Functions for DLPack Speed Exchange and Stream Handling (#…

Kathryn-cat Dec 4, 2025

77b90b7

Revert "Re-enable torch.compile tests for Python 3.12 and Windows (#1…

pytorchmergebot Dec 4, 2025

9ff16b8

Updated hypot to accept cpu scalar Tensor even on GPU. (#169302)

cleonard530 Dec 4, 2025

5a9322c

Remove unnecessary code in exception registration (#169365)

cyyever Dec 4, 2025

b74e823

Remove outdated Makefile targets (#164679)

cyyever Dec 4, 2025

b5e5ee9

improve accuracy of division by scalar (#169507)

ngimel Dec 4, 2025

6211bfd

[Inductor] Fix combo kernels for cpu backend (#167781)

karthickai Dec 4, 2025

607a535

[effects] Various effect/dce fixes (#169141)

angelayi Dec 4, 2025

e7a86c4

[CI] Pin CPython to 3.14.0 (#169592)

malfet Dec 4, 2025

3199a3e

localtensor: test_utils.py (#169532)

dolpm Dec 4, 2025

d2b0d81

Add check_lowerbound config for AOTI lowering (#169430)

ColinPeppler Dec 5, 2025

7143efb

Remove outdated skip conditons of CUDA and ROCm (#166391)

cyyever Dec 5, 2025

4380129

[DTensor] Fix slow sharding prop for stack (#169519)

wconstab Dec 3, 2025

6be8f42

Revert "[dynamo] Refactor isinstance(x, ConstantVariable) to x.is_pyt…

pytorchmergebot Dec 5, 2025

ff43f43

[Profiler][PrivateUse1] Fix ProfilerState typo ('Disable'→'Disabled')…

KarhouTam Dec 5, 2025

64f29c0

[dynamo][dicts] Prepare for dict refactor for fewer conflicts (#169589)

anijain2305 Dec 4, 2025

1a8de02

Fix log_sigmoid_backward_batch_rule on XPU (#169215)

Silv3S Dec 5, 2025

8c13e8b

Revert getAllOperatorsFor changes (#167860, #162218) (#169281)

ezyang Dec 1, 2025

6470af7

[dynamo] Prep for dict key refactor - add hash related methods to VT …

anijain2305 Dec 4, 2025

3a3156f

Skip NVIDIA cleanup on CPU runners after #169431 (#169625)

huydhn Dec 5, 2025

5213a72

Enable ruff SIM115 check (#169437)

cyyever Dec 5, 2025

d914547

Make TORCH_BOX support const ref arguments (#169563)

mikaylagawarecki Dec 4, 2025

9f7c46b

[Dynamo][Guard]Add the user-friendly TYPE_MATCH for type (#169025)

fxdawnn Dec 4, 2025

79e8d6a

[12/N] Use Python 3.10 typing (#169355)

cyyever Dec 5, 2025

2169e66

Revert "[Inductor] ReLU/GELU(Addmm) fusions (#168157)"

pytorchmergebot Dec 5, 2025

053cbc1

[dynamo] assign random IDs to graph break registry (#169617)

williamwen42 Dec 4, 2025

240780a

[dynamo] add new graph break registry entries via linter to random sp…

williamwen42 Dec 4, 2025

365a6c8

[BE] Upgrade XPU support package to 2025.3 (#166829)

chuanqi129 Dec 5, 2025

e36b4d6

[3/N] Remove unused header inclusion (#169200)

cyyever Dec 5, 2025

0ff77ef

[DTensor] unbacked matmuls for no-redistribute case (#168051)

pianpwk Dec 4, 2025

8b683e5

[BE] Delete `install_vision` from Docker builds (#169609)

malfet Dec 5, 2025

4cd5afa

[vllm hash update] update the pinned vllm hash (#165274)

pytorchupdatebot Dec 5, 2025

a573c49

[Inductor] Add debug output for specific pattern matching (#169603)

karthickai Dec 4, 2025

17ec1c3

[Inductor] Fix combo kernels by populating constants for equal_to_1 a…

karthickai Dec 4, 2025

5f76830

[ROCm] unskip some ROCm inductor UTs (#169564)

jataylo Dec 5, 2025

7375582

Fix slotscheck warnings (#169348)

cyyever Dec 5, 2025

09f1800

Revert "[BE] Delete `install_vision` from Docker builds (#169609)"

pytorchmergebot Dec 5, 2025

a0a2ae6

[Dynamo][Guards]Fix TLParse CPP guard message with sorting get_leaf_g…

fxdawnn Dec 5, 2025

6de6685

Use context managers (#169447)

cyyever Dec 5, 2025

3158c04

[Inductor] Fix pattern matcher FailedMatch format string (#169611)

karthickai Dec 4, 2025

716edc3

Triton 3.6 update. Add CooperativeReductionTests to slow (#169630)

atalman Dec 5, 2025

d064b22

activation offloading implementation (#167880)

xuanzhang816 Dec 5, 2025

eca37f0

activation offloading reordering for comp<>comm overlaps (#168316)

xuanzhang816 Dec 5, 2025

a113483

[precompile] disable dispatch when deepcloning in PrecompileContext.r…

bobrenjc93 Dec 5, 2025

a70d81a

[pytorch] Add env variable to enable IPC for expandable segments (#16…

pavanbalaji Dec 5, 2025

82e30f3

Fix viable strict update when no green commit found (#169585)

clee2000 Dec 5, 2025

ae64a53

make the float4 dtype support equality comparisons (#169575)

vkuzo Dec 4, 2025

3cf2f19

add copy_ support for float4 dtype (#169595)

vkuzo Dec 4, 2025

fa21963

[MPS] Add input/indices shape validation for MaxUnpool{1,2,3}d (#169261)

lingebeng Dec 5, 2025

3620149

Correct some grammatical and expression errors in the CONTRIBUTING.md…

wenlinchong17-web Dec 5, 2025

4710fd9

Remove unnecessary uses of thrust::pair (#168941)

cyyever Dec 5, 2025

b6e3d44

Merge remote-tracking branch 'upstream/main' into develop_IFU_20251205

github-actions[bot] Dec 5, 2025