[QNN EP] Add BFloat16 dtype support in QNN EP #26987

tirupath-qti · 2026-01-12T22:56:48Z

Description

QNN NPU backend supports BFloat16 dtype for many operators
QNN EP adds a new session option "htp_bf16_enable" to enable Users to signal processing the Float32 graph in BFloat16 precision
When User specifies "htp_bf16_enable", the QNN EP lowers incoming Float32 Ort graph into BFloat16 QNN graph.
The ORT CPU fallback still receives Float32 partitions.
The lowered QNN graph still accepts float32 inputs, outputs and constant initializers. The QNN EP inserts Cast operators to do the necessary precision switch.

Motivation and Context

This enables computing accuracy sensitive float32 models in bfloat16 precision on Qualcomm NPU accelerator to improve inference time w.r.t computing in float32 precision.

- QNN NPU backend supports BFloat16 dtype for many operators - QNN EP adds a new session option "htp_bf16_enable" to enable Users to signal processing the Float32 graph in BFloat16 precision - When User specifies "htp_bf16_enable", the QNN EP lowers incoming Float32 Ort graph into BFloat16 QNN graph. - The ORT CPU fallback still receives Float32 partitions. - The lowered QNN graph still accepts float32 inputs, outputs and constant initializers. The QNN EP inserts Cast operators to do the necessary precision switch.

tianleiwu · 2026-01-13T19:34:46Z

From AI

Summary

This PR introduces BFloat16 (BF16) support to the QNN Execution Provider, specifically for the HTP backend. Currently, this is opt-in via a new session option htp_bf16_enable. When enabled, the EP "lowers" a Float32 graph to use BF16 precision on the NPU, effectively treating Float32 operations as BF16.

Key Changes

Session Option: htp_bf16_enable added to control the feature.
Graph Translation: QnnModelWrapper updated to convert Float32 tensors/operations to BF16 when constructing the QNN graph.
Boundary Handling: Helper logic inserts necessary casts so that inputs/outputs remain Float32 compatible (for CPU fallback or interface consistency).
Testing: Comprehensive bf16_handling_test.cc added.

Review Analysis

Correctness

Approach: The "lowering" approach (running FP32 graph as BF16) is a common technique for accelerators. The implementation correctly handles this by intercepting the graph construction rather than rewriting the ONNX model file.
Boundary Safety: It's crucial that CPU fallback nodes receive FP32. The PR mentions this is handled ("ORT CPU fallback still receives Float32 partitions").
Constants: Handling constant initializers (converting them to BF16) is critical for memory saving and is included.

Performance

Throughput: BF16 on NPU should offer significant speedups over FP32 (if FP32 was even supported efficiently) and arguably better accuracy/ease-of-use than INT8 quantization for some models.
Memory: Storing weights in BF16 cuts memory usage by half compared to FP32.

Conclusion

This feature bridges a gap for users needing higher precision than INT8 but better performance than FP32. The implementation via a session option is safe and non-intrusive. LGTM.

yuslepukhin · 2026-01-13T20:29:24Z

/azp run Linux QNN CI Pipeline, Win_TRT_Minimal_CUDA_Test_CI, Windows ARM64 QNN CI Pipeline, Windows GPU CUDA CI Pipeline, Windows GPU DML CI Pipeline, Windows GPU Doc Gen CI Pipeline, Windows GPU TensorRT CI Pipeline, Windows OpenVINO CI Pipeline, Windows x64 QNN CI Pipeline

azure-pipelines · 2026-01-13T20:29:44Z

Azure Pipelines successfully started running 4 pipeline(s).

yuslepukhin · 2026-01-13T21:27:19Z

/mnt/vss/_work/1/s/onnxruntime/test/providers/qnn/bf16_handling_test.cc: In function ‘void onnxruntime::test::RunBF16ModelTest(const GetTestModelFn&, const std::vector&, onnxruntime::test::ExpectedEPNodeAssignment, int, float)’:
/mnt/vss/_work/1/s/onnxruntime/test/providers/qnn/bf16_handling_test.cc:51:58: error: unused parameter ‘input_shape’ [-Werror=unused-parameter]
51 | const std::vector<int64_t>& input_shape,
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~
/mnt/vss/_work/1/s/onnxruntime/test/providers/qnn/bf16_handling_test.cc: At global scope:
/mnt/vss/_work/1/s/onnxruntime/test/providers/qnn/bf16_handling_test.cc:50:13: error: ‘void onnxruntime::test::RunBF16ModelTest(const GetTestModelFn&, const std::vector&, onnxruntime::test::ExpectedEPNodeAssignment, int, float)’ defined but not used [-Werror=unused-function]
50 | static void RunBF16ModelTest(const GetTestModelFn& build_test_case,
| ^~~~~~~~~~~~~~~~
/mnt/vss/_work/1/s/onnxruntime/test/providers/qnn/bf16_handling_test.cc:39:23: error: ‘onnxruntime::test::GetTestModelFn onnxruntime::test::BuildBF16ConvTestCase(const onnxruntime::test::TestInputDef&, const onnxruntime::test::TestInputDef&)’ defined but not used [-Werror=unused-function]
39 | static GetTestModelFn BuildBF16ConvTestCase(const TestInputDef& input_def,
| ^~~~~~~~~~~~~~~~~~~~~
/mnt/vss/_work/1/s/onnxruntime/test/providers/qnn/bf16_handling_test.cc:28:23: error: ‘onnxruntime::test::GetTestModelFn onnxruntime::test::BuildBF16MatMulTestCase(const onnxruntime::test::TestInputDef&, const onnxruntime::test::TestInputDef&)’ defined but not used [-Werror=unused-function]
28 | static GetTestModelFn BuildBF16MatMulTestCase(const TestInputDef& input1_def,
| ^~~~~~~~~~~~~~~~~~~~~~~
/mnt/vss/_work/1/s/onnxruntime/test/providers/qnn/bf16_handling_test.cc:17:23: error: ‘onnxruntime::test::GetTestModelFn onnxruntime::test::BuildBF16AddTestCase(const onnxruntime::test::TestInputDef&, const onnxruntime::test::TestInputDef&)’ defined but not used [-Werror=unused-function]
17 | static GetTestModelFn BuildBF16AddTestCase(const TestInputDef& input1_def,
| ^~~~~~~~~~~~~~~~~~~~

Copilot

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

onnxruntime/core/providers/qnn/builder/qnn_model_wrapper.cc

onnxruntime/test/providers/qnn/bf16_op_test.cc

- Ensure that bf16 only works when soc model is 88 or above - Handling initializers in QNN that are not part of ONNX to be converted to bfloat16 from float32. - Implemented a RAII guard for RestoreFp32AfterValidation

adrianlizarraga · 2026-01-14T19:47:19Z

/azp run Linux QNN CI Pipeline, Win_TRT_Minimal_CUDA_Test_CI, Windows ARM64 QNN CI Pipeline, Windows GPU Doc Gen CI Pipeline

azure-pipelines · 2026-01-14T19:47:38Z

Azure Pipelines successfully started running 4 pipeline(s).

qti-ashwshan · 2026-01-14T21:33:14Z

Looks like there were a couple warnings on the test file and failing linux pipeline. Added a patch for it.

adrianlizarraga · 2026-01-14T21:42:59Z

/azp run Linux QNN CI Pipeline, Win_TRT_Minimal_CUDA_Test_CI, Windows ARM64 QNN CI Pipeline, Windows GPU Doc Gen CI Pipeline

azure-pipelines · 2026-01-14T21:43:19Z

Azure Pipelines successfully started running 4 pipeline(s).

qti-ashwshan · 2026-01-14T21:49:22Z

Rebased main to get latest fix for test failures

adrianlizarraga · 2026-01-14T21:53:07Z

/azp run Linux QNN CI Pipeline, Win_TRT_Minimal_CUDA_Test_CI, Windows ARM64 QNN CI Pipeline, Windows GPU Doc Gen CI Pipeline

azure-pipelines · 2026-01-14T21:53:26Z

Azure Pipelines successfully started running 4 pipeline(s).

onnxruntime/test/providers/qnn/bf16_op_test.cc

adrianlizarraga · 2026-01-14T23:47:17Z

/azp run Linux QNN CI Pipeline, Win_TRT_Minimal_CUDA_Test_CI, Windows ARM64 QNN CI Pipeline, Windows GPU Doc Gen CI Pipeline

azure-pipelines · 2026-01-14T23:47:47Z

Azure Pipelines successfully started running 4 pipeline(s).

### Description - QNN NPU backend supports BFloat16 dtype for many operators - QNN EP adds a new session option "htp_bf16_enable" to enable Users to signal processing the Float32 graph in BFloat16 precision - When User specifies "htp_bf16_enable", the QNN EP lowers incoming Float32 Ort graph into BFloat16 QNN graph. - The ORT CPU fallback still receives Float32 partitions. - The lowered QNN graph still accepts float32 inputs, outputs and constant initializers. The QNN EP inserts Cast operators to do the necessary precision switch. ### Motivation and Context - This enables computing accuracy sensitive float32 models in bfloat16 precision on Qualcomm NPU accelerator to improve inference time w.r.t computing in float32 precision. --------- Co-authored-by: Ashwath Shankarnarayan <ashwshan@qti.qualcomm.com>

### Description - QNN NPU backend supports BFloat16 dtype for many operators - QNN EP adds a new session option "htp_bf16_enable" to enable Users to signal processing the Float32 graph in BFloat16 precision - When User specifies "htp_bf16_enable", the QNN EP lowers incoming Float32 Ort graph into BFloat16 QNN graph. - The ORT CPU fallback still receives Float32 partitions. - The lowered QNN graph still accepts float32 inputs, outputs and constant initializers. The QNN EP inserts Cast operators to do the necessary precision switch. ### Motivation and Context - This enables computing accuracy sensitive float32 models in bfloat16 precision on Qualcomm NPU accelerator to improve inference time w.r.t computing in float32 precision. --------- Co-authored-by: Ashwath Shankarnarayan <ashwshan@qti.qualcomm.com> (cherry picked from commit 530a1fb)

### Description This PR cherry-picks the following changes for the 1.24.0 release. ### Cherry-picked Commits | Commit | Commit Title | Author | |---|---|---| | 744e7fe | Add type definitions, registration, utilities for INT2/UINT2 support (#26824) | vraspar | | 530a1fb | [QNN EP] Add BFloat16 dtype support in QNN EP (#26987) | tirupath-qti | | 8e050d1 | Implement new experimental lookup-based matrix multiplication method(TMAC) (#26695) | vraspar | | 2d2ba6b | [MLAS/CPU EP] Improve performance of Silu activation path within the QuickGelu CPU kernel (#26753) | Hariharan Seshadri | | 1c02b79 | [QNN EP] Add support for handling 0-dimension for Concat Op (#27000) | Ashwath Shankarnarayan | | cc2b01b | Fix ClipQuantFusion crash when Clip has multiple input edges (#27016) | Edward Chen | | bbd3850 | [QNN EP] Support quantized BatchNorm with per-channel DQ params on QNN HTP (#26959) | qti-yuduo | | d8f0318 | Add API to get ep graph partitioning info (#26781) | Adrian Lizarraga | | b912b18 | [OVEP] OpenVINO EP Features and bug-fixes for ORT-1.24 - Follow up (#27007) | Preetha Veeramalai | | ba11af4 | [QNN-EP] Add MatMulNBits translation for GPU (#26340) | quic-tirupath | | c03c419 | [MLAS/NEON] Add dedicated kernel for depthwise convolution for ARM64 using NEON intrinsics (#26688) | Hariharan Seshadri | | e7dfd69 | [QNN-EP] Support alternate Layernorm fusion pattern in QNN preprocess (#26060) | qti-mattsinc | | 4013dc1 | Implement multithreading in qgemm_kleidi (#26301) | Melike Kaptan | | 9f06181 | [CXX] Enable users to specify custom OrtSyncStream via RunOptions (#26988) | Dmitri Smirnov | | cfccd64 | Added support for QMX kernels in MLAS (#26849) | qti-vaiskv | | 29d9b2f | Tweak external resource importer handle structs (#27040) | Scott McKay | | 9d108d0 | [QNN EP] Add QuickGELU operator support for QNN provider (#27034) | tirupath-qti | | b35688f | Add INT2 and UINT2 support for QDQ, transpose and cast ops (#27022) | vraspar | | 6d34aba | Introducing BF16 Pointwise NCHWc Convolution for Arm64 (#26838) | Rohanjames1997 | | 36017ad | [EP ABI] Add CreateCustomOpDomains() API for plugin EP to register custom ops (#27050) | Chi Lo | | 50a03e4 | Add a new pipeline for CUDA 13 nuget builds (#27023) | eserscor | | a0d4439 | [EP ABI] Update Graph_GetGraphView() implementation (#26711) | Chi Lo | | 34bb209 | [webgpu] Fix a bug for im2col (#27069) | Wenqin Yang | | 46e8d45 | [QNN EP] Add FusedMatMul operator support (#27044) | tirupath-qti | | 5e7e7a3 | Disable Float32_2Bits_Asymmetric_256x256 test (#27046) | vraspar | | 39f966e | Fix Doxygen documentation build error in onnxruntime_c_api.h (#27083) | Nick Eubanks | | 8a7a797 | Print tensor for new packed type of 2 bits (#27064) | Tianlei Wu | | 01f40e6 | Fix GPU JAR testing on Linux (#27011) | eserscor | | b6ed7f3 | Fix warning around ununsed code in QNN Android Emulator builds by clang (#27026) | Hariharan Seshadri | | d7daa45 | Raise the timeout for the ios simulator job (#27045) | Hariharan Seshadri | | 7e1d818 | upgrade emsdk to 4.0.23 (#27029) | Yulong Wang | | 347b990 | Fix failing mainline build on Arm64 linux (#27101) | Rohanjames1997 | | f481b17 | Add dedicated API to support extracting compatibility string from model metadata (#27015) | adrastogi | --------- Signed-off-by: Liqun Fu <liqun.fu@microsoft.com> Signed-off-by: bfilipek <bartlomiej.filipek@intel.com> Signed-off-by: dependabot[bot] <support@github.com> Signed-off-by: Jonathan Clohessy <jonathan.clohessy@arm.com> Signed-off-by: Christian Bourjau <christian.bourjau@quantco.com> Signed-off-by: melkap01 <melike.kaptan@arm.com> Co-authored-by: vraspar <vrajang@outlook.com> Co-authored-by: tirupath-qti <tirupath@qti.qualcomm.com> Co-authored-by: Ashwath Shankarnarayan <ashwshan@qti.qualcomm.com> Co-authored-by: Liqun Fu <liqun.fu@microsoft.com> Co-authored-by: carzh <wolfivyaura@gmail.com> Co-authored-by: Hector Li <hecli@microsoft.com> Co-authored-by: carzh <carolinezhu@microsoft.com> Co-authored-by: Vrajang Parikh <vrparikh@microsoft.com> Co-authored-by: Hariharan Seshadri <shariharan91@gmail.com> Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Co-authored-by: Edward Chen <18449977+edgchen1@users.noreply.github.com> Co-authored-by: Yuduo Wu <yuduow@qti.qualcomm.com> Co-authored-by: Adrian Lizarraga <adlizarraga@microsoft.com> Co-authored-by: Preetha Veeramalai <preetha.veeramalai@intel.com> Co-authored-by: jatinwadhwa921 <110383850+jatinwadhwa921@users.noreply.github.com> Co-authored-by: jatinwadhwa921 <jatin.wadhwa@intel.com> Co-authored-by: saurabh <saurabh1.kale@intel.com> Co-authored-by: Ankit Maheshkar <ankit.maheshkar@intel.com> Co-authored-by: sfatimar <sahar.fatima@intel.com> Co-authored-by: Javier Martinez <javier.e.martinez@intel.com> Co-authored-by: Bartlomiej Filipek <bartlomiej.filipek@intel.com> Co-authored-by: bopeng1234 <bo.peng@intel.com> Co-authored-by: Eric Crawford <eric.r.crawford@intel.com> Co-authored-by: MayureshV1 <47039074+MayureshV1@users.noreply.github.com> Co-authored-by: TejalKhade28 <tejal.khade@intel.com> Co-authored-by: Vishnudas Thaniel S <vishnudas.thaniel.s@intel.com> Co-authored-by: Yaru Du <yaru.du@intel.com> Co-authored-by: Ryan Metcalfe <107415876+RyanMetcalfeInt8@users.noreply.github.com> Co-authored-by: Dvoretckii, Mikhail <mikhail.dvoretckii@intel.com> Co-authored-by: Pallavi Gupta <pallavi.gupta@intel.com> Co-authored-by: Jianhui Dai <jianhui.j.dai@intel.com> Co-authored-by: Jiajia Qin <jiajiaqin@microsoft.com> Co-authored-by: Changming Sun <chasun@microsoft.com> Co-authored-by: Fei Chen <feich@microsoft.com> Co-authored-by: Yulong Wang <7679871+fs-eire@users.noreply.github.com> Co-authored-by: Akupadhye <aupadhye@qti.qualcomm.com> Co-authored-by: Wang Ning <ning4.wang@intel.com> Co-authored-by: Maximilian Müller <44298237+gedoensmax@users.noreply.github.com> Co-authored-by: Chi Lo <54722500+chilo-ms@users.noreply.github.com> Co-authored-by: George Wu <jywu@microsoft.com> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: Wanming Lin <wanming.lin@intel.com> Co-authored-by: quic-calvnguy <quic_calvnguy@quicinc.com> Co-authored-by: Jie Chen <jie.a.chen@intel.com> Co-authored-by: xhcao <xinghua.cao@intel.com> Co-authored-by: Wei-Sheng Chin <wschin@outlook.com> Co-authored-by: quic-hungjuiw <quic_hungjuiw@quicinc.com> Co-authored-by: Ian Hunter <ianfhunter@gmail.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: kunal-vaishnavi <115581922+kunal-vaishnavi@users.noreply.github.com> Co-authored-by: Jeff Kilpatrick <jkilpatrick@qti.qualcomm.com> Co-authored-by: Jeff Kilpatrick <jkilpat@qti.qualcomm.com> Co-authored-by: Scott McKay <skottmckay@gmail.com> Co-authored-by: Nenad Banfic <46795300+nenad1002@users.noreply.github.com> Co-authored-by: derdeljan-msft <derdeljan@microsoft.com> Co-authored-by: n1harika <niharika.sathish@intel.com> Co-authored-by: Ryan Metcalfe <ryan.metcalfe@intel.com> Co-authored-by: Jaswanth Gannamaneni <jaswanth.gannamaneni@intel.com> Co-authored-by: Klimenko, Mikhail <mikhail.klimenko@intel.com> Co-authored-by: liang <gxgaoliang@126.com> Co-authored-by: Garth Long <garth.long@intel.com> Co-authored-by: Jonathan Clohessy <jonathan.clohessy@arm.com> Co-authored-by: Akshay Sonawane <111780983+apsonawane@users.noreply.github.com> Co-authored-by: Christopher Warrington <chwarr@microsoft.com> Co-authored-by: Ishwar Raut <iraut@nvidia.com> Co-authored-by: Gaurav Garg <gaugarg@nvidia.com> Co-authored-by: Xinpeng Dou <15529241576@163.com> Co-authored-by: adrastogi <aditya.rastogi@microsoft.com> Co-authored-by: Aditya Rastogi <adityar@ntdev.microsoft.com> Co-authored-by: qti-hungjuiw <hungjuiw@qti.qualcomm.com> Co-authored-by: Pradeep Sakhamoori <psakhamoori@microsoft.com> Co-authored-by: Adam Pocock <adam.pocock@oracle.com> Co-authored-by: mingyue <131847423+mingyueliuh@users.noreply.github.com> Co-authored-by: Susanta Bhattacharjee <susanta.bhattacharjee@intel.com> Co-authored-by: Jozef Wludzik <jozef.wludzik@intel.com> Co-authored-by: Rajeev Sekar <rajeevsekar21@gmail.com> Co-authored-by: Mayuresh M Varerkar <mayuresh.m.varerkar@intel.com> Co-authored-by: Copilot <198982749+Copilot@users.noreply.github.com> Co-authored-by: Wenqin Yang <wenqin.yang@intel.com> Co-authored-by: xieofxie <xieofxie@126.com> Co-authored-by: hualxie <hualxie@microsoft.com> Co-authored-by: Joshua Lochner <admin@xenova.com> Co-authored-by: Christian Bourjau <cbourjau@users.noreply.github.com> Co-authored-by: Xiaofei Han <xiaofeihan@microsoft.com> Co-authored-by: Dmitri Smirnov <yuslepukhin@users.noreply.github.com> Co-authored-by: chunghow-qti <chunghow@qti.qualcomm.com> Co-authored-by: Guenther Schmuelling <guschmue@microsoft.com> Co-authored-by: Jiawei Shao <jiawei.shao@intel.com> Co-authored-by: czekun <chen.zekun@intel.com> Co-authored-by: Jaskaran Singh Nagi <jaskaran.singh.nagi@intel.com> Co-authored-by: quic-tirupath <quic_tirupath@quicinc.com> Co-authored-by: qti-mattsinc <mattsinc@qti.qualcomm.com> Co-authored-by: Melike Kaptan <melike.kaptan@arm.com> Co-authored-by: Damien Dooley <damien.dooley@arm.com> Co-authored-by: qti-vaiskv <vaiskv@qti.qualcomm.com> Co-authored-by: Rohanjames1997 <rohan.james4@gmail.com> Co-authored-by: eserscor <erscor@microsoft.com> Co-authored-by: eserscor <247253654+eserscor@users.noreply.github.com> Co-authored-by: Nick Eubanks <nieubank@microsoft.com> Co-authored-by: adrastogi <8368026+adrastogi@users.noreply.github.com> Co-authored-by: Rohanjames1997 <rohanjms@amazon.com>

- QNN NPU backend supports BFloat16 dtype for many operators - QNN EP adds a new session option "htp_bf16_enable" to enable Users to signal processing the Float32 graph in BFloat16 precision - When User specifies "htp_bf16_enable", the QNN EP lowers incoming Float32 Ort graph into BFloat16 QNN graph. - The ORT CPU fallback still receives Float32 partitions. - The lowered QNN graph still accepts float32 inputs, outputs and constant initializers. The QNN EP inserts Cast operators to do the necessary precision switch. - This enables computing accuracy sensitive float32 models in bfloat16 precision on Qualcomm NPU accelerator to improve inference time w.r.t computing in float32 precision. --------- Co-authored-by: Ashwath Shankarnarayan <ashwshan@qti.qualcomm.com>

yuslepukhin requested a review from Copilot January 13, 2026 20:28

Copilot started reviewing on behalf of yuslepukhin January 13, 2026 20:29 View session

yuslepukhin added the ep:QNN issues related to QNN exeution provider label Jan 13, 2026

Copilot AI reviewed Jan 13, 2026

View reviewed changes

adrianlizarraga reviewed Jan 14, 2026

View reviewed changes

[QNN EP] Add BFloat16 dtype support in QNN EP

a32388c

- Ensure that bf16 only works when soc model is 88 or above - Handling initializers in QNN that are not part of ONNX to be converted to bfloat16 from float32. - Implemented a RAII guard for RestoreFp32AfterValidation

- Silence unused function warnings

c51a481

qti-ashwshan added 2 commits January 14, 2026 13:46

Merge branch 'master' of github.com:CodeLinaro/onnxruntime into HEAD

1aaa827

Merge branch 'master' of github.com:CodeLinaro/onnxruntime into HEAD

1da6179

adrianlizarraga reviewed Jan 14, 2026

View reviewed changes

onnxruntime/test/providers/qnn/bf16_op_test.cc Show resolved Hide resolved

- unused variable input_shape

3a8c274

adrianlizarraga approved these changes Jan 15, 2026

View reviewed changes

tianleiwu enabled auto-merge (squash) January 15, 2026 18:27

tianleiwu merged commit 530a1fb into microsoft:main Jan 15, 2026
88 checks passed

edgchen1 added the release:1.24.0 label Jan 15, 2026

This was referenced Jan 21, 2026

1.24.0 release cherry-pick round 1 #27103

Closed

1.24.0 release cherry-pick round 1 #27104

Merged

tianleiwu removed the release:1.24.0 label Jan 23, 2026

[QNN EP] Add BFloat16 dtype support in QNN EP #26987

[QNN EP] Add BFloat16 dtype support in QNN EP #26987

Uh oh!

Conversation

tirupath-qti commented Jan 12, 2026

Description

Motivation and Context

Uh oh!

tianleiwu commented Jan 13, 2026

From AI

Summary

Key Changes

Review Analysis

Correctness

Performance

Conclusion

Uh oh!

yuslepukhin commented Jan 13, 2026

Uh oh!

azure-pipelines bot commented Jan 13, 2026

Uh oh!

yuslepukhin commented Jan 13, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

adrianlizarraga commented Jan 14, 2026

Uh oh!

azure-pipelines bot commented Jan 14, 2026

Uh oh!

qti-ashwshan commented Jan 14, 2026

Uh oh!

adrianlizarraga commented Jan 14, 2026

Uh oh!

azure-pipelines bot commented Jan 14, 2026

Uh oh!

qti-ashwshan commented Jan 14, 2026

Uh oh!

adrianlizarraga commented Jan 14, 2026

Uh oh!

azure-pipelines bot commented Jan 14, 2026

Uh oh!

Uh oh!

adrianlizarraga commented Jan 14, 2026

Uh oh!

azure-pipelines bot commented Jan 14, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants