165 Pull requests merged by 42 people

• Fix link in using-executorch-building-from-source.md

  #9756 merged Mar 31, 2025

• [Release 0.6] Update doc getting started

  #9748 merged Mar 31, 2025

• [Build] use timespec_get to get current time

  #5252 merged Mar 31, 2025

• Add ET_PRI_* macros for Tensor methods that differ with ATen

  #9768 merged Mar 31, 2025

• Serialize NamedDataStoreOutput into PTD.

  #9758 merged Mar 31, 2025

• Support non-default dim orders in pybindings + aten bridge

  #9736 merged Mar 31, 2025

• Arm backend: Refactor any, bitwise, logical tests

  #9499 merged Mar 31, 2025

• Arm backend: Check memory allocation on target

  #9735 merged Mar 31, 2025

• Qualcomm AI Engine Direct - Replace private variable with function

  #8724 merged Mar 31, 2025

• Introduce missing APIs to lower ExportedProgram objects directly

  #8909 merged Mar 31, 2025

• Update mimi export test

  #9755 merged Mar 29, 2025

• Remove old tokenizer/ directory in ExecuTorch

  #9728 merged Mar 29, 2025

• Add bf16 support to unary_ufunc_realh

  #9599 merged Mar 29, 2025

• Serialize NamedDataStoreOutput into PTD.

  #9125 merged Mar 29, 2025

• [Release 0.6] update CI pins

  #9746 merged Mar 29, 2025

• [Bug Fix] Fix padding when running in NHWC

  #9729 merged Mar 28, 2025

• Improve ET_CHECK_OR_RETURN_FALSE error messages in kernels/portable

  #9751 merged Mar 28, 2025

• Pin cmake version < 4.0.0

  #9750 merged Mar 28, 2025

• Openvino backend for Executorch to enable inference on Intel CPUs, GPUs, NPUs

  #8573 merged Mar 28, 2025

• [Release 0.6] update version.txt

  #9744 merged Mar 28, 2025

• Deprioritize top level -O2 in CMAKE_CXX_FLAGS_RELEASE

  #9394 merged Mar 28, 2025

• Arm backend: Add support to eq.Scalar

  #9715 merged Mar 28, 2025

• Fix CoreML pybinding module

  #9724 merged Mar 28, 2025

• Arm backend: Make download of ethos-u dependencies optional

  #9734 merged Mar 28, 2025

• Improve android related docs

  #9725 merged Mar 28, 2025

• Pin cmake version < 4.0.0

  #9732 merged Mar 28, 2025

• Add a convenient constructor

  #9707 merged Mar 28, 2025

• Add mixed dtype check for XNNPACK partitioner

  #9612 merged Mar 27, 2025

• Revert "Depend on extension/threadpool, not thread_parallel_interface, in buck (#9511)"

  #9727 merged Mar 27, 2025

• Arm backend: Clean up shift support

  #9573 merged Mar 27, 2025

• Move ExecutorchRuntimeBridge into fb dir.

  #9722 merged Mar 27, 2025

• Support alpha in scalar add/sub cases

  #9703 merged Mar 27, 2025

• convert to .md and clean up content on main docs page

  #9663 merged Mar 27, 2025

• [ET-VK] Adding repeat support to add_copy_packed_dim_offset_node function.

  #9721 merged Mar 27, 2025

• [ET-VK][ez] Small fix to SPIR-V compilation artifact caching

  #9720 merged Mar 27, 2025

• Move ExecutorchRuntimeBridge into fb dir.

  #9704 merged Mar 27, 2025

• [ET-VK] Adding repeat support to add_copy_packed_dim_offset_node function.

  #9414 merged Mar 27, 2025

• [ET-VK][ez] Small fix to SPIR-V compilation artifact caching

  #9716 merged Mar 27, 2025

• set -x in build_apple_frameworks.sh

  #9326 merged Mar 27, 2025

• Arm backend: Add MobileNet v3 testcase

  #9223 merged Mar 27, 2025

• Arm backend: Support scalar_tensor op

  #9711 merged Mar 27, 2025

• Arm backend: Use gitlab and bump version of ethos-u-core-driver to 25.02

  #9708 merged Mar 27, 2025

• [ET-VK] Cache compiled SPIR-V and only recompile when changed

  #9701 merged Mar 27, 2025

• Remove exception fall back on checkpoint loading

  #9660 merged Mar 27, 2025

• Add method variants in dynamic shim to use core tensor/evalue/result types

  #7834 merged Mar 27, 2025

• Arm backend: Add missing __init__.py to passes

  #9710 merged Mar 27, 2025

• Arm backend: Only describe pytest and not python unittest in README.md

  #9709 merged Mar 27, 2025

• Support channels_last format in portable upsample kernels

  #9526 merged Mar 27, 2025

• Arm backend: Add tosa_spec info to .tosa files

  #9392 merged Mar 27, 2025

• Module API to execute with a single Tensor.

  #9706 merged Mar 27, 2025

• Use different bundle to lookup resources for SwiftPM.

  #9702 merged Mar 27, 2025

• Add Value equality API.

  #9700 merged Mar 27, 2025

• [ET-VK] Cache compiled SPIR-V and only recompile when changed

  #9652 merged Mar 27, 2025

• Handle more Value types in Module.

  #9699 merged Mar 27, 2025

• Add more types to create a Value from.

  #9698 merged Mar 27, 2025

• More tests for scalar constructor.

  #9697 merged Mar 27, 2025

• Overloads for scalar constructor.

  #9696 merged Mar 27, 2025

• Tensor constructor to create with a single scalar.

  #9695 merged Mar 27, 2025

• More tests for scalars constructor.

  #9694 merged Mar 27, 2025

• Overloads for scalars constructor.

  #9693 merged Mar 27, 2025

• Tensor constructor to create with an array of scalars.

  #9692 merged Mar 27, 2025

• Overloads for data Tensor constructor.

  #9691 merged Mar 27, 2025

• Tensor constructor to create with an existing NSData.

  #9690 merged Mar 27, 2025

• Add Module forward API with overloads.

  #9689 merged Mar 27, 2025

• Overloads for Module execute API.

  #9688 merged Mar 27, 2025

• Add Module execute API.

  #9687 merged Mar 27, 2025

• Add Module methodName API.

  #9686 merged Mar 27, 2025

• Add Module loadMethod API.

  #9685 merged Mar 27, 2025

• Add Module load API.

  #9684 merged Mar 27, 2025

• Create Module with a filepath and load mode.

  #9683 merged Mar 27, 2025

• Create Value with Tensor.

  #9682 merged Mar 27, 2025

• Value default initializer.

  #9681 merged Mar 27, 2025

• Add Tensor equality API.

  #9680 merged Mar 27, 2025

• Add Tensor resizing API.

  #9679 merged Mar 27, 2025

• Make Tensor conform to NSCopying protocol.

  #9678 merged Mar 27, 2025

• Tensor constructor to create with an existing Tensor.

  #9677 merged Mar 27, 2025

• Tensor accessorts to get raw data buffer.

  #9676 merged Mar 27, 2025

• Overloads for the bytes Tensor constrcutor.

  #9675 merged Mar 27, 2025

• Tensor constructor to create with a raw pointer by copying the data.

  #9674 merged Mar 27, 2025

• Tensor helpers to get the size of a data type and number of elements of a shape.

  #9673 merged Mar 27, 2025

• Overloads for the bytesNoCopy Tensor constrcutor.

  #9672 merged Mar 27, 2025

• [ET-VK][ez] Fix printing shader IRs

  #9666 merged Mar 27, 2025

• [ET-VK][benchmarking][ez] Don't perform copies when benchmarking

  #9671 merged Mar 27, 2025

• Tensor constructor to create with a raw pointer w/o copying.

  #9670 merged Mar 27, 2025

• [ET-VK][benchmarking][ez] Don't perform copies when benchmarking

  #9468 merged Mar 27, 2025

• Tensor properties to get shape, etc.

  #9668 merged Mar 27, 2025

• Use llama_inputs config option only if available

  #9667 merged Mar 27, 2025

• Initialize Tensor with native TensorPtr.

  #9646 merged Mar 27, 2025

• [ET-VK][ez] Fix printing shader IRs

  #9648 merged Mar 26, 2025

• [ExecuTorch] Arm Ethos: Do not depend on torch.testing._internal

  #8929 merged Mar 26, 2025

• [Executorch][llama] Change runner to decouple prompt length from sequence length

  #9650 merged Mar 26, 2025

• [ET-VK][ez] Fix 8 bit linear compute shader dispatch

  #9664 merged Mar 26, 2025

• [BE][ET-VK] update max_pool2d to use layout gen

  #9657 merged Mar 26, 2025

• [BE][ET-VK] update full to use layout gen

  #9656 merged Mar 26, 2025

• [BE][ET-VK] update clone to use layout gen

  #9655 merged Mar 26, 2025

• [BE][ET-VK] update permute to use layout gen

  #9654 merged Mar 26, 2025

• Test CoreML support when building wheels

  #9647 merged Mar 26, 2025

• [ET-VK][ez] Fix 8 bit linear compute shader dispatch

  #9531 merged Mar 26, 2025

• Upgrade arm runner for phi4-mini

  #9658 merged Mar 26, 2025

• Remove tombstone messages proactively

  #8703 merged Mar 26, 2025

• Update pinned optimum-executorch commit hash in the CI

  #9653 merged Mar 26, 2025

• Deprecate non-internal elementwise_util APIs

  #9621 merged Mar 26, 2025

• elementwise_util: s/common/compute/ almost everywhere and deprecate SAME_AS_COMPUTE

  #9613 merged Mar 26, 2025

• Depend on extension/threadpool, not thread_parallel_interface, in buck

  #9511 merged Mar 26, 2025

• Fix -Wglobal-constructors/-Wshadow for executor_runner targets

  #9509 merged Mar 26, 2025

• add sample input for new core op elu.out

  #9577 merged Mar 26, 2025

• [BE][ET-VK] update max_pool2d to use layout gen

  #9591 merged Mar 26, 2025

• [BE][ET-VK] update full to use layout gen

  #9590 merged Mar 26, 2025

• [BE][ET-VK] update clone to use layout gen

  #9589 merged Mar 26, 2025

• [BE][ET-VK] update permute to use layout gen

  #9588 merged Mar 26, 2025

• Add portable ELU implementation + test

  #9520 merged Mar 26, 2025

• Updates docs to reflect changes in CoreML

  #9645 merged Mar 26, 2025

• Aot compiler fix

  #9634 merged Mar 26, 2025

• Enable CoreML by default on macOS wheel builds

  #9483 merged Mar 26, 2025

• Alternative approach to handling memory offset shift

  #9406 merged Mar 26, 2025

• Update docs after moving Android ExecuTorchDemo app

  #9601 merged Mar 26, 2025

• Qualcomm AI Engine Direct - Mimi Enablement Stage 1

  #9570 merged Mar 26, 2025

• [Executorch][llama] Change runner to decouple prompt length from sequence length

  #9594 merged Mar 26, 2025

• Arm backend: set scalar div xfails to non strict.

  #9644 merged Mar 26, 2025

• Arm backend: Extend convolution support check to 3d

  #9640 merged Mar 26, 2025

• Add buck target for hf_download

  #9603 merged Mar 26, 2025

• Add export_llama performance regression test using expected ops

  #9158 merged Mar 26, 2025

• Arm backend: Set qtol=1 for flaky tests.

  #9641 merged Mar 26, 2025

• Arm backend: fix test_llama.py

  #9575 merged Mar 26, 2025

• Update Qualcomm SoCs support list

  #9592 merged Mar 26, 2025

• Turn on EXECUTORCH_BUILD_PYBIND when implicitly wanted

  #9611 merged Mar 26, 2025

• Revert "Bump HF version"

  #9600 merged Mar 25, 2025

• Enum for Tensor scalar type and shape dynamism.

  #9609 merged Mar 25, 2025

• Enum for Value tag.

  #9608 merged Mar 25, 2025

• Enum for Module load mode.

  #9606 merged Mar 25, 2025

• Utility helpers to convert between std::vector and NSArray.

  #9597 merged Mar 25, 2025

• Rename deduceScalarType to deduceType

  #9604 merged Mar 25, 2025

• Remove usage of environment variables in wheel building

  #9583 merged Mar 25, 2025

• Switch to new ao quant api for 8da4w (#8501)

  #8772 merged Mar 25, 2025

• Reset Dynamo at Setup for all tests

  #9561 merged Mar 25, 2025

• Fix conflicting pytorch nightly versions

  #9593 merged Mar 25, 2025

• Utility helper to extract a value from NSNumber.

  #9595 merged Mar 25, 2025

• Remove Android ExecuTorch demo

  #9554 merged Mar 25, 2025

• Utility helper to deduce scalar type from NSNumber.

  #9552 merged Mar 25, 2025

• Add a test target to SwiftPM package.

  #9557 merged Mar 25, 2025

• Bump HF version

  #9408 merged Mar 25, 2025

• Create intermediate directories in ETCoreMLModelManager

  #9562 merged Mar 25, 2025

• Refactor serialize.py

  #9579 merged Mar 25, 2025

• Create a CoreML inmemoryfs pybinding target

  #9482 merged Mar 25, 2025

• Tiny fixes to make gcc pedantic build happy

  #8933 merged Mar 25, 2025

• Add NamedData to flat_tensor schema

  #9571 merged Mar 25, 2025

• Arm backend: Add ComputeConstantOpsAOT pass

  #9504 merged Mar 25, 2025

• Refactor serialize.py

  #9124 merged Mar 25, 2025

• Add NamedData to flat_tensor schema

  #9123 merged Mar 25, 2025

• Download checkpoints from HuggingFace

  #9538 merged Mar 25, 2025

• Add missing etdump dependency

  #9547 merged Mar 25, 2025

• Make a real buck file for Apple extension to run tests

  #9560 merged Mar 25, 2025

• Rename phi-4-mini to phi_4_mini

  #9539 merged Mar 25, 2025

• Separate CoreML util and inmemoryfs targets to be reusable

  #9481 merged Mar 24, 2025

• Bump torchao pin to enable dynamic shapes in lowbit

  #9555 merged Mar 24, 2025

• Refactoring memory planning to allow running multiple algorithms

  #8440 merged Mar 24, 2025

• Fix exir.load/save to handle named data store map

  #9485 merged Mar 24, 2025

• Bug fix in Ethos runtime backend

  #9517 merged Mar 24, 2025

• Add Phi-4-mini README.md

  #9302 merged Mar 24, 2025

• Fix qwen import typo

  #9535 merged Mar 24, 2025

• Remove dead test

  #9363 merged Mar 24, 2025

• Use cached PyTorch wheels on MacOS jobs

  #9484 merged Mar 24, 2025

• Add SmolLM (smollm2)

  #9541 merged Mar 24, 2025

• Android aar update and enable emulator test

  #9524 merged Mar 24, 2025

• Updates torchao pin to enable shared embedding quantization

  #9548 merged Mar 24, 2025

46 Pull requests opened by 25 people

• fix rsqrt test case

  #9556 opened Mar 24, 2025

• Fix instrumentation test

  #9564 opened Mar 25, 2025

• Make a real buck file for Apple extension to run tests

  #9566 opened Mar 25, 2025

• Arm backend: Size Test for Cortex-M

  #9569 opened Mar 25, 2025

• Added hardtanh operator

  #9574 opened Mar 25, 2025

• Bump PyTorch pin to 2025-03-25

  #9578 opened Mar 25, 2025

• Add SupportedTensorDtypes::{BOOL,REALH}

  #9584 opened Mar 25, 2025

• relax tolerances for all unary float ops

  #9585 opened Mar 25, 2025

• NOCOMMIT: PoC: vectorize just expm1

  #9586 opened Mar 25, 2025

• Use sccache to accelerate android build

  #9587 opened Mar 25, 2025

• Update test_ios_ci.sh to use the app from executorch-examples repo

  #9596 opened Mar 25, 2025

• quantize_ decouple weight quant and dequant output dtypes

  #9642 opened Mar 26, 2025

• Qualcomm AI Engine Direct - Add QNN support for to_edge_transform_and_lower

  #9643 opened Mar 26, 2025

• executorch

  #9661 opened Mar 26, 2025

• decode to prefill

  #9662 opened Mar 26, 2025

• Fix build gating on optimized_portable_kernels

  #9665 opened Mar 26, 2025

• [wip] Make extension_module static

  #9669 opened Mar 27, 2025

• Arm backend: update docs to new template

  #9712 opened Mar 27, 2025

• Check for null method in module execute

  #9713 opened Mar 27, 2025

• Convert Args to DictConfig (#9450)

  #9717 opened Mar 27, 2025

• Add README.md for the NXP eIQ Neutron Backend

  #9718 opened Mar 27, 2025

• Back out "Add method variants in dynamic shim to use core tensor/evalue/result types"

  #9719 opened Mar 27, 2025

• Update install_android.sh

  #9723 opened Mar 27, 2025

• Unbreak CI job failures caused by cmake upgrade

  #9730 opened Mar 28, 2025

• Fix Fake bench data type

  #9731 opened Mar 28, 2025

• Remove unecessary tests from pull

  #9738 opened Mar 28, 2025

• set INTERFACE_LINK_LIBARIES for extension_threadpool

  #9739 opened Mar 28, 2025

• Add build_optimized_size_test.sh

  #9740 opened Mar 28, 2025

• Migrate elementwise_util callers to the variants with out_dtypes in template arguments

  #9741 opened Mar 28, 2025

• Save some size in dtype_util when dtype selective build is not in use

  #9742 opened Mar 28, 2025

• smart_ptr multi-turn demo

  #9743 opened Mar 28, 2025

• etdumpfilter impl

  #9752 opened Mar 28, 2025

• [XNNPACK] Serialize weights as fp16 rather than fp32

  #9753 opened Mar 29, 2025

• use kUnsetDebugHandle for unset debug handle in etdump delegation log

  #9754 opened Mar 29, 2025

• Validate that method inputs have the expected dim order

  #9757 opened Mar 29, 2025

• Fix breaks from googletest upgrade

  #9760 opened Mar 30, 2025

• Error handling log intermediate output delegate

  #9761 opened Mar 31, 2025

• Refactor and add ET_EXPECT_DEATH tests to LogDelegateIntermediateOutput

  #9762 opened Mar 31, 2025

• Add cadence.where_Scalar op

  #9764 opened Mar 31, 2025

• [ET-VK] Store weights transposed for int8 linear

  #9765 opened Mar 31, 2025

• [ET-VK] Efficient tiled int8 matmul

  #9766 opened Mar 31, 2025

• Move replacement of conv to linear/fc to opt level 2

  #9769 opened Mar 31, 2025

• Update pybinding README

  #9770 opened Mar 31, 2025

• Turn on XNNPACK by default for wheels

  #9773 opened Mar 31, 2025

• Remove redundant install_executorch.sh call in QNN CI job

  #9774 opened Mar 31, 2025

• Bump swiftpm packages version

  #9775 opened Mar 31, 2025

24 Issues closed by 12 people

• Failed to resize output tensor for XNNExecutor. CALL_DELEGATE execute failed at instruction 6: 0x10

  #9749 closed Mar 31, 2025

• Standalone AAR build script

  #8261 closed Mar 29, 2025

• QNN: mobilebert failed to generate Qnn context binary

  #7946 closed Mar 28, 2025

• Duplicate Dynamic Quant Chain in XNNPACK Lowering Flow

  #9022 closed Mar 28, 2025

• Partitioner to not delegate mixed dtype operators

  #9023 closed Mar 27, 2025

• Getting Started Path and Streamlined Flow for Android + Java

  #7883 closed Mar 27, 2025

• Populate "Using ExecuTorch on Android" doc page

  #8536 closed Mar 27, 2025

• Provide an example model in the getting started guide

  #8283 closed Mar 27, 2025

• Provide API to query backend availability at runtime

  #8267 closed Mar 27, 2025

• A simple install command gives everything a user need

  #7460 closed Mar 27, 2025

• Top level documentation navigation revamp

  #8391 closed Mar 27, 2025

• Create delegate documentation template

  #8269 closed Mar 27, 2025

• Fill out XNNPACK backend quantization docs

  #8526 closed Mar 27, 2025

• Move ExecuTorchDemo Android app to pytorch-labs/executorch-demo

  #8789 closed Mar 27, 2025

• Distribute Android binaries through a Maven repository

  #8259 closed Mar 27, 2025

• Typo in import path of examples/models/qwen2_5/__init__.py

  #9525 closed Mar 27, 2025

• Migrate export_llama to new ao quantize API

  #8422 closed Mar 27, 2025

• ARM ML embedded evaluation Kit support

  #7423 closed Mar 27, 2025

• Has any one check the executorch implementation on arm (arm_executor_runner.cpp) with mobilenetv2 and real image ?

  #7568 closed Mar 27, 2025

• ./install_executorch.sh fails on pip due to dependency conflicts

  #9445 closed Mar 26, 2025

• exir.load/exir.save is broken after weight sharing (named data store) addition to EdgeProgramManager

  #9365 closed Mar 25, 2025

• Using Hugging Face model card name in export_llama

  #8872 closed Mar 25, 2025

• [Request help] torch._dynamo.exc.TorchRuntimeError about indices and SymInt

  #9486 closed Mar 25, 2025

• Refactor memory planning to support running multiple algorithms and choose the one that has the best output results.

  #8466 closed Mar 24, 2025

40 Issues opened by 19 people

• Return "platform not supported" when using PyTorch on intel-based Macbooks

  #9772 opened Mar 31, 2025

• [Request Impl] Apply Segment Serialization in Bundled Program

  #9771 opened Mar 31, 2025

• Cannot run .pte generated with Python in C++

  #9763 opened Mar 31, 2025

• Error on import protable_lib via pybindings

  #9745 opened Mar 28, 2025

• [MediaTek] mtk_neuron compile error: 3D permutation not support

  #9733 opened Mar 28, 2025

• export fails with latest executorch

  #9714 opened Mar 27, 2025

• Does SpinQuant implemented R3 when using quantized kv cache?

  #9705 opened Mar 27, 2025

• Remove unused parameters before serializing

  #9659 opened Mar 26, 2025

• Remove the requirement to install requirements for CoreML

  #9651 opened Mar 26, 2025

• Handle 4-bit CoreML affine quantization via PT2E

  #9649 opened Mar 26, 2025

• Why does the weight data type of the Linear layer become FP32 during the runtime when load fp16.pte (fp16 Llama 3.2-1B model)?

  #9639 opened Mar 26, 2025

• QNN Export Llama Model Error "failed to load context binary"

  #9637 opened Mar 26, 2025

• Facing linking error "undefined reference to `executorch::extension::FileDataLoader::from"

  #9636 opened Mar 26, 2025

• Support enumerated shapes for models with multiple inputs

  #9632 opened Mar 25, 2025

• Improve CoreML model loading time

  #9631 opened Mar 25, 2025

• Support better tiling algorithms for ANE

  #9630 opened Mar 25, 2025

• Improve GPU performance

  #9629 opened Mar 25, 2025

• CI to cover different iOS versions

  #9628 opened Mar 25, 2025

• BC/FC guarantees clearly stated

  #9627 opened Mar 25, 2025

• Establish a process for monitoring and resolving CoreML GH issues, including an SLA

  #9626 opened Mar 25, 2025

• Continuous coverage and performance monitoring of GA models for CoreML backend.

  #9625 opened Mar 25, 2025

• Clean up ET CoreML documentation

  #9624 opened Mar 25, 2025

• Support RuntimeSpec

  #9623 opened Mar 25, 2025

• Enumerated shape support

  #9622 opened Mar 25, 2025

• Support dynamic shape in outputBacking

  #9620 opened Mar 25, 2025

• Support reshapeFrequency runtime spec

  #9619 opened Mar 25, 2025

• Dynamic shape support

  #9618 opened Mar 25, 2025

• CoreML quantization is supported by PyTorch flows (PT2E).

  #9617 opened Mar 25, 2025

• Windows build failure: Failed to query buck for sources.

  #9616 opened Mar 25, 2025

• A supported ANE-friendly llama model with good out-of-the-box performance and accuracy.

  #9615 opened Mar 25, 2025

• All GA models are supported by CoreML and have similar performance to using CoreML via torch.export

  #9614 opened Mar 25, 2025

• All GA models export to CoreML via. torch.export with the same performance and user experience as torch.jit.trace

  #9610 opened Mar 25, 2025

• ARM baremetal tests are not running properly

  #9602 opened Mar 25, 2025

• Llava runner job failing after HF transformers update

  #9598 opened Mar 25, 2025

• Beefing up CONTRIBUTING.md to lower the barrier of entry for external contributors

  #9582 opened Mar 25, 2025

• Verify Mimi's accuracy and performance

  #9581 opened Mar 25, 2025

• Export and run lora llama3b with adapter weights

  #9580 opened Mar 25, 2025

• Missing .so lib for training pybinding

  #9576 opened Mar 25, 2025

• [Request help] Failed running call_function <built-in function getitem>

  #9567 opened Mar 25, 2025

• `from executorch.exir import CaptureConfig` fails with ExecuTorch pip installed in editable mode

  #9558 opened Mar 24, 2025

71 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Qualcomm AI Engine Direct - Add submodule quant config setting

  #9355 commented on Mar 31, 2025 • 7 new comments

• [Proposal] Support Multiple Prefill + Decode in a loop

  #9466 commented on Mar 26, 2025 • 6 new comments

• Arm backend: Add ethos_u_minimal_example jupyter notebook

  #9543 commented on Mar 31, 2025 • 4 new comments

• Arm backend: Tosa tools update

  #9451 commented on Mar 31, 2025 • 2 new comments

• RFC: Specialize for non-mixed-dtype in elementwise_util

  #9388 commented on Mar 28, 2025 • 2 new comments

• Implement unary_ufunc functions using elementwise_util

  #9386 commented on Mar 28, 2025 • 2 new comments

• elementwise_util: don't cast the result of compute_fun back to the common type

  #9385 commented on Mar 28, 2025 • 2 new comments

• Add MediaTek backend CI

  #7120 commented on Mar 27, 2025 • 1 new comment

• Add (and fix) OSS check that et_operator_library dtype_selective_build builds

  #9245 commented on Mar 27, 2025 • 0 new comments

• Support llama3 training

  #9149 commented on Mar 31, 2025 • 0 new comments

• Add im2row per tensor overload

  #9121 commented on Mar 31, 2025 • 0 new comments

• Don't ignore parallel_for result in op_sdpa

  #9119 commented on Mar 29, 2025 • 0 new comments

• Use external hf_tokenizer in llama runner

  #9112 commented on Mar 26, 2025 • 0 new comments

• Arm backend: Added 8 new unit tests for testing various passes.

  #9037 commented on Mar 28, 2025 • 0 new comments

• Arm backend: Move ReplaceScalarTensorWithFullPass to transforms

  #8998 commented on Mar 31, 2025 • 0 new comments

• Arm backend: Remove unused github ci git setup

  #8833 commented on Mar 31, 2025 • 0 new comments

• Adding new memory plannig algorithm heap_optimized_greedy

  #8773 commented on Mar 28, 2025 • 0 new comments

• (WIP) Neuron buffer allocator decouple from ExecuTorch framework

  #8760 commented on Mar 27, 2025 • 0 new comments

• Support multi-module exports in Inspector and ETRecord

  #8336 commented on Mar 31, 2025 • 0 new comments

• Fix ODR violation due to const-mismatch with eigens gemm definition.

  #8049 commented on Mar 31, 2025 • 0 new comments

• Update Codegen.cmake for improved Python package path handling

  #6730 commented on Mar 31, 2025 • 0 new comments

• QNN GPU or DSP backend issue

  #5914 commented on Mar 26, 2025 • 0 new comments

• Add arm64 size_test

  #9252 commented on Mar 27, 2025 • 0 new comments

• Qualcomm AI Engine Direct - oss model enablement (EfficientSAM)

  #9266 commented on Mar 26, 2025 • 0 new comments

• Add support for torch.pow in the Arm backend

  #9309 commented on Mar 31, 2025 • 0 new comments

• Arm backend: Enable mypy lintrunner for backends/arm/test

  #9357 commented on Mar 31, 2025 • 0 new comments

• Arm backend: Make passes preserve and update node metadata

  #9362 commented on Mar 31, 2025 • 0 new comments

• Make PyTorch headers available in optimized_portable_kernels, define ET_USE_PYTORCH_HEADERS

  #9384 commented on Mar 28, 2025 • 0 new comments

• Refactor elementwise_util: create variants with out_dtypes in template argument list

  #9387 commented on Mar 28, 2025 • 0 new comments

• Arm backend: Use dbg_fail when node visitors raise exceptions

  #9391 commented on Mar 31, 2025 • 0 new comments

• BroadcastIndexesRange: leading 1s don't require true broadcasting

  #9431 commented on Mar 28, 2025 • 0 new comments

• Add vectorization in elementwise_util

  #9432 commented on Mar 31, 2025 • 0 new comments

• Arm Backend: Add New Silu and SDPA Decomp passes to arm_pass_manager

  #9448 commented on Mar 25, 2025 • 0 new comments

• Add optimized ELU implementation

  #9521 commented on Mar 28, 2025 • 0 new comments

• Refactor LLMEdgeManager's to_dtype

  #9542 commented on Mar 25, 2025 • 0 new comments

• STASH: randn_out

  #9553 commented on Mar 24, 2025 • 0 new comments

• Is the QNN backend support the model of Llama 3.2 3B instead of XNNPACK？

  #9311 commented on Mar 26, 2025 • 0 new comments

• Refactor binary op partitioner configs under binary op config class

  #9024 commented on Mar 26, 2025 • 0 new comments

• [RFC] Canonicalize python file structure

  #8699 commented on Mar 25, 2025 • 0 new comments

• [Android] Error running a transformer decoder-based neural network

  #8665 commented on Mar 25, 2025 • 0 new comments

• Error in pip editable mode in export_llama

  #9278 commented on Mar 25, 2025 • 0 new comments

• ConstraintViolationError

  #6782 commented on Mar 25, 2025 • 0 new comments

• Parallelize portable ops if threadpool is available, with fallback to parallel_for-as-for-loop

  #8932 commented on Mar 25, 2025 • 0 new comments

• SM8750 Htp not added in Android apk building

  #8454 commented on Mar 25, 2025 • 0 new comments

• Recipe and Input class definition for executorch.export API

  #9366 commented on Mar 25, 2025 • 0 new comments

• CoreML palletization works in PT2E flow

  #9518 commented on Mar 24, 2025 • 0 new comments

• double free or corruption (out) on running Llava on x86_64

  #4869 commented on Mar 24, 2025 • 0 new comments

• Move Int8DynamicActivationIntxWeightConfig out of torchao/experimental.

  #9516 commented on Mar 24, 2025 • 0 new comments

• [etLLM] New config system to export_llama

  #9449 commented on Mar 24, 2025 • 0 new comments

• Migrate backend quantizers/partitioners to use new torchao quant primitives

  #9513 commented on Mar 24, 2025 • 0 new comments

• Document end-to-end quantization journey in ExecuTorch

  #9515 commented on Mar 24, 2025 • 0 new comments

• I want to understand the allocations made on STM32s by Executorch

  #9491 commented on Mar 24, 2025 • 0 new comments

• Enable dynamic shapes for torchao:8daXw quant ops

  #8981 commented on Mar 24, 2025 • 0 new comments

• [pytorch hash update] update the pinned pytorch hash

  #4589 commented on Mar 31, 2025 • 0 new comments

• Binary Size: Selective Build

  #9010 commented on Mar 31, 2025 • 0 new comments

• Vectorize optimized_portable_ops versions of portable ops?

  #9241 commented on Mar 31, 2025 • 0 new comments

• Update Arm Ethos U backend docs using backend template

  #8530 commented on Mar 31, 2025 • 0 new comments

• Error running .pte model with executor_runner

  #8923 commented on Mar 31, 2025 • 0 new comments

• _load_for_executorch pybinding cannot load model for training

  #4908 commented on Mar 31, 2025 • 0 new comments

• log_intermediate_output_delegate cannot handle error as expected

  #9549 commented on Mar 30, 2025 • 0 new comments

• Missing try-before-set tests for datasink in etdumpgen tests

  #9551 commented on Mar 30, 2025 • 0 new comments

• ExecuTorch Scalar to() supports fewer types than c10::Scalar, breaking source compatibility

  #9500 commented on Mar 29, 2025 • 0 new comments

• [v0.6.0] Release Tracker

  #9253 commented on Mar 28, 2025 • 0 new comments

• Support exporting QNN models with Python wheels out-of-the-box

  #9474 commented on Mar 28, 2025 • 0 new comments

• [RFC] Objective-C/Swift APIs for ExecuTorch

  #8360 commented on Mar 27, 2025 • 0 new comments

• Expose runtime error codes / results in JNI

  #8260 commented on Mar 27, 2025 • 0 new comments

• torch._check_is_size is not being recognized by CoreML partitioner

  #9213 commented on Mar 27, 2025 • 0 new comments

• Llama 3.2 3B Core ML poor output quality

  #9395 commented on Mar 27, 2025 • 0 new comments

• CoreML model works with torch.jit.trace, but not torch.export.export

  #9506 commented on Mar 27, 2025 • 0 new comments

• (release/0.5) ./install_requirements.sh fails due to conflicting dependencies.

  #9307 commented on Mar 26, 2025 • 0 new comments

• Build pip wheel and validate AOT workflow on different backends & host environments in CI

  #9019 commented on Mar 26, 2025 • 0 new comments