Google Git
Sign in
chromium/external/github.com/google-ai-edge/LiteRT/HEAD
  1. 5bb05f7 Add graph fusion. by Fengwu Yao · 5 hours ago upstream/main
  2. 36f3ed0 Merge pull request #9709 from google-ai-edge/litert_pr_974747642 by jl45G · 7 hours ago
  3. b4e4443 Add TFLiteConverter.from_mlir_bytecode in tf.lite Python API. by Majid Dadashi · 8 hours ago
  4. 7536bca Add TFLiteConverter.from_mlir_bytecode in tf.lite Python API. by Majid Dadashi · 8 hours ago
  5. 9f7aea8 Update readme as cc APIs are now header only by Google AI Edge · 8 hours ago
  6. 3b7022f pass TensorRT artifacts through JIT handles by Google AI Edge · 9 hours ago
  7. d416805 share TensorRT weights across engine bytecode by Google AI Edge · 9 hours ago
  8. 9693097 Refactored compiler plugin and dispatch api to allow AOT compilation for tensor rt engine bytecode by Google AI Edge · 9 hours ago
  9. f82cd3b Introduced customized gemv kernel for faster decode speed by Google AI Edge · 9 hours ago
  10. 9638c1e Copybara import of the project: by destro4evr · 10 hours ago
  11. 428d279 optimize the sdpa kernel. by Fengwu Yao · 10 hours ago
  12. d1122fa Clean up JNI target labels. by Google AI Edge · 10 hours ago
  13. aeb2f05 Support madvise for GPU weight conversion in LiteRT GPU delegate by Terry Heo · 10 hours ago
  14. 41c7a93 Synchronize LiteRT Objective-C error codes and tensor element types by Terry Heo · 10 hours ago
  15. b80f93a Fix copyright header format for Copybara license validation by Google AI Edge · 10 hours ago
  16. f4ae858 Update Intel copyright headers to remove (C) by Jun Jiang · 10 hours ago
  17. bb43853 Various IrModel fixes by Grant Jensen · 10 hours ago
  18. d8d6636 Reverts 71dd84403db7b9485a61c9747093d0a53955b1a2 by Grant Jensen · 10 hours ago
  19. f777a1b Merge pull request #9621 from akodanka:ov_version_update by Copybara-Service · 10 hours ago
  20. 40e5a1e This is an automatic update to a device compatibility allowlist. by Alexandru Paiu · 12 hours ago
  21. cc2519b Fix tensorflow/lite/delegates/ynnpack/attention_bench build in open-source. by Volodymyr Kysenko · 13 hours ago
  22. 0ec5dd0 Add converter pass debugging and crash reproduction infrastructure by Majid Dadashi · 24 hours ago
  23. 46df83e Add range check to a few StableHLO ops by Changming Sun · 25 hours ago
  24. 4aa7a0e Add range check to a few StableHLO ops by Changming Sun · 25 hours ago
  25. 0249059 Set fuzzing test main to fuzztest_gtest_main for LiteRT OSS export. by Changming Sun · 26 hours ago
  26. 68a67c8 Enable QSV and QDQ quantization pipeline in StableHLO-to-TFLite converter by Majid Dadashi · 30 hours ago
  27. e8b1ea4 Add fuzzing tests back by Changming Sun · 31 hours ago
  28. e43bce2 PR #109378: cmake: Exclude subdirectories from all builds by tdarote · 32 hours ago
  29. 64cfcb8 Fix uninitialized pointer crash in DispatchDelegate initialization by Andrew Zhang · 35 hours ago
  30. 84acdee Fix heap buffer overflow in DynamicUpdateSliceInt4 by clamping memcpy length. by Google AI Edge · 2 days ago
  31. bc3afe2 Make it possible to look up quantized embeddings directly. by Quentin Khan · 2 days ago
  32. 4479f07 Copybara import of the project: by destro4evr · 2 days ago
  33. 68fb2ea Fix unaligned external weight serialization in flatbuffer export by Google AI Edge · 2 days ago
  34. 66a81e7 Fix unaligned external weight serialization in flatbuffer export by Google AI Edge · 2 days ago
  35. 9ce1a07 Fix unaligned external weight serialization in flatbuffer export by Google AI Edge · 2 days ago
  36. dc32e93 Enable batched GPU weight conversion on the OpenCL delegate backend by Terry Heo · 3 days ago chromium/8038
  37. 71dd844 If the IrModel creates intermediate tensors w/ non float type, allow the SMM to create tensors of that type instead of forcing them into float. by Grant Jensen · 4 days ago
  38. 5020ea1 Update maven bulid script to include litert support maven package by Google AI Edge · 4 days ago
  39. 1c2309e Update litert support build script by Google AI Edge · 4 days ago
  40. d72d902 Declare litert support api and litert support with different package names by Google AI Edge · 4 days ago
  41. 609a0b8 follow up on flatbuffer schema by Google AI Edge · 4 days ago
  42. 40b9f9c Fix flatbuffer schema generation conflict by Google AI Edge · 4 days ago
  43. e0302bd Rename propagate-qsv pass to propagate-qparams and add VHLO quant custom call legalization by Majid Dadashi · 4 days ago
  44. 25824ae Rename propagate-qsv pass to propagate-qparams and add VHLO quant custom call legalization by Majid Dadashi · 4 days ago
  45. acd98e9 Merge pull request #7596 from Rjasuja:npu_hal_intgr by Copybara-Service · 4 days ago
  46. 69c0561 Add Mean Generator to LiteRT ATS by Gerardo Carranza · 4 days ago
  47. 4fe5c66 Rollback of PR #126109 by Changming Sun · 4 days ago
  48. 87f68a9 Fix TFLite FlatBuffer SignatureDef indexing and low-bit constant export by Majid Dadashi · 4 days ago
  49. 8908c11 Fix TFLite FlatBuffer SignatureDef indexing and low-bit constant export by Majid Dadashi · 4 days ago
  50. 79ce93a Support grouped transposed convolutions in LiteRT-Torch and StableHLO legalization. by Google AI Edge · 4 days ago
  51. fa3db6b Rollback of PR #126177 by Google AI Edge · 5 days ago
  52. e56b2ab Bump the OpenVINO version to 2026.3.1 (build 22476) by Anoob Anto Kodankandath · 5 days ago
  53. 1084539 Make OpenVINO package version a single generated config by Anoob Anto Kodankandath · 5 days ago
  54. a9254e9 Update leakr configuration for support lib oss by Google AI Edge · 5 days ago
  55. d3f2eb1 Add support for reading and writing Data in Swift TensorBuffer. by Jun Jiang · 5 days ago
  56. fd031ae Remove resolve_symbols_in_exec build configuration from LiteRT. by Chunlei Niu · 5 days ago
  57. 30574ad Remove libandroid dependency from CMake build by Terry Heo · 5 days ago
  58. 8d9d021 PR #126162: Security: fix heap out-of-bounds read in TFLite TRANSPOSE (permutation not validated) by Elshad Toklayev · 5 days ago
  59. 9fe5be4 PR #126177: Security: fix heap out-of-bounds read in TFLite SLICE (Prepare skips bounds validation) by Elshad Toklayev · 5 days ago
  60. 9e12a6f Enable TFLite Bazel and CMake presubmit by Google AI Edge · 5 days ago
  61. 2b596d1 PR #126109: Add an OSS-Fuzz target for the TensorFlow Lite interpreter by Elshad Toklayev · 5 days ago
  62. cf28033 PR #125809: Fix the CMake build of the Metal delegate by Cocoa · 5 days ago
  63. 108efbe PR #123818: lite: fix memory leak in TFLiteSavedModelConverterV2.convert() by DakshPrajapati-afk · 6 days ago
  64. 2b3136a Migrate to absl/status/status_macros.h by Juhyun Lee · 6 days ago
  65. 47683b8 PR #122082: Fix signed-integer-overflow UB in TFLite optimized integer Add kernel. by Wei Wang · 6 days ago
  66. 0a25a8d PR #119633: docs: fix typographical and spelling errors across MLIR, TFLite, and Python ops by Saketh Reddy Pingili · 6 days ago
  67. eaa4e58 PR #119633: docs: fix typographical and spelling errors across MLIR, TFLite, and Python ops by Saketh Reddy Pingili · 6 days ago
  68. f75d71a Add Swift bindings for LiteRT CPU options. by Jun Jiang · 6 days ago
  69. 672c96e Reverts 679762a6cfa8c707716e53efd96ce45ce5068993 by Grant Jensen · 6 days ago
  70. e4d8d77 Fuse input reshape into cache update op by Byungchul Kim · 6 days ago
  71. fcf84ff PR #125807: fix: compare strcmp against 0 when matching GPU delegate option keys by Cocoa · 6 days ago
  72. bf524ee Merge pull request #7656 from neuropilot-captain:260528_upstream by Copybara-Service · 6 days ago
  73. 5587c36 Validate int16 kernel parameters during prepare. by Changming Sun · 6 days ago
  74. cb237dd Disable sanitizers and heap check for LiteRT OpenCL and custom tensor buffer tests by Terry Heo · 6 days ago
  75. 1205d65 Package the LiteRT Metal accelerator dylib into Swift xcframeworks. by Jun Jiang · 6 days ago
  76. 78676a6 Change LockedBufferSpan to make it possible to create multiple spans from the same locked object. by Quentin Khan · 6 days ago
  77. e2f83ba Merge pull request #9455 from gunes-arm:pr/add-arm-dependencies by Copybara-Service · 6 days ago
  78. 0b3d622 Internal changes only. by Fengwu Yao · 6 days ago
  79. 7c00791 Copybara import of the project: by destro4evr · 6 days ago
  80. fc117d0 Branch tflite_support to litert/support_libs by Google AI Edge · 6 days ago
  81. 15b8c0c Internal changes only. by Fengwu Yao · 6 days ago
  82. a391ee1 Internal changes only. by Fengwu Yao · 6 days ago
  83. 71bd09f Add SwiGLU composite op by Fengwu Yao · 6 days ago
  84. 7c7410a Add Metal unit tests for LiteRT Objective-C compiled model by Terry Heo · 6 days ago
  85. 8c59081 [IrModel] Expand Resize dtype coverage by Grant Jensen · 6 days ago
  86. ca3b70e [IrModel] Populate shared weights tensor w/ placeholder in emb lookup by Grant Jensen · 6 days ago
  87. 06a7739 Add a comment pointing users to the officially supported Tensor SoCs documentation. by Google AI Edge · 7 days ago
  88. 679762a If the IrModel creates intermediate tensors w/ non float type, allow the SMM to create tensors of that type instead of forcing them into float. by Grant Jensen · 7 days ago
  89. b478b3b [IrModel] Support scalar axis tensor for Split op by Grant Jensen · 7 days ago
  90. a8e5840 [IrModel] Support scalar pack inputs by Grant Jensen · 7 days ago
  91. bad370e PR #125814: Make TfLiteGpuDelegateV2Delete and TFLGpuDelegateDelete accept nullptr by Cocoa · 7 days ago
  92. 37e8b22 OpenVINO NPU: Route NPU inference through Android NPU HAL for job priority by Ritul Jasuja · 7 days ago
  93. 928830d Remove hyphens from package names by Gunes Bayir · 7 days ago
  94. 569e161 Internal change by Google AI Edge · 7 days ago
  95. d3d87e3 Add Float16 support to TFLite NEG operator. by Weiyi Wang · 7 days ago
  96. 6146e4a Update tflite patterns to respect resource attributes by Majid Dadashi · 7 days ago
  97. 174302b Add an interface for loading a model from mlir bytecode by Majid Dadashi · 7 days ago
  98. e75e633 Add an `:auto` special value to derive the weight cache path from the model file. by Quentin Khan · 7 days ago
  99. c9c37e5 Stop forcing constant tensor sharing when a weight loader is provided by Lynne Jiang · 7 days ago
  100. 5407d29 PR #125667: Add init to MLModelConfiguration by Amarjot Gill · 7 days ago