Google Git
Sign in
chromium/external/github.com/google-ai-edge/LiteRT/HEAD
  1. 4d96d04 Refactor LiteRtCompilerPluginPartition for better modularity and readability by Google AI Edge · 6 hours ago upstream/main
  2. 2ebed3f PR #125646: Make TFLite conversion bindings safe for free-threaded Python by Mina Ibrahim · 8 hours ago
  3. 8213ec0 PR #124109: Fix TFLite FuseMulAndFullyConnected on shape-expanding Mul by PrabinDevkota · 9 hours ago
  4. b01b8ed Automated Code Change by Google AI Edge · 13 hours ago
  5. 6dace2a Enable --enable-externalize-weights option in litert lm builder by Google AI Edge · 13 hours ago
  6. 10100f1 Add full TFLite Bazel and CMake build presubmit by Google AI Edge · 14 hours ago
  7. cb1ffba Preserve reduced TFLite converter surface in OSS exports by Google AI Edge · 15 hours ago
  8. a2d2ca7 Preserve reduced TFLite surface in OSS exports by Google AI Edge · 15 hours ago
  9. 9823021 Preserve reduced TFLite surface in OSS exports by Google AI Edge · 15 hours ago
  10. 6a51150 Move the `safetensor_loader` to the `utils` folder. by Quentin Khan · 19 hours ago
  11. 21738ed This is an internal change by Google AI Edge · 21 hours ago
  12. fabad75 Update cmake build script to use the correct path in github by Google AI Edge · 21 hours ago
  13. 3456e95 Explicitly avoid using absl in public cc headers by Google AI Edge · 21 hours ago
  14. 042b965 Back @litertjs/core Tensor objects with LiteRtTensorHandle by Google AI Edge · 22 hours ago
  15. 1865174 PR #120316: Spelling: Fix various typos in comments by Oliver Zhang · 27 hours ago
  16. cf53e6f Use AHardwareBuffer to allocate memory on Mali by Marissa Ikonomidis · 28 hours ago
  17. f674577 Move `perfetto_session` to the `utils` folder. by Quentin Khan · 4 days ago
  18. 23a3670 Document host memory buffer alignment requirement in LiteRT by Terry Heo · 4 days ago
  19. 932787a Fix out-of-bounds read in TFLite BIDIRECTIONAL_SEQUENCE_LSTM by Adrian Ciausu · 4 days ago
  20. 61f6415 Bump LiteRT Converter version to 0.5.0. by Google AI Edge · 4 days ago
  21. 7d683c3 Only use `-Wl,-lc++abi` for dispatch API shared libraries in OSS. by Jordan Rupprecht · 4 days ago
  22. 3c64443 This is an internal change by Google AI Edge · 4 days ago
  23. 7a29494 Clean up dependencies. by Quentin Khan · 4 days ago
  24. 0d94449 Internal change only. by Google AI Edge · 4 days ago
  25. a7c881c Refactor ATS operator registration by splitting ats.cc into modular files. by Gerardo Carranza · 4 days ago
  26. ffa15af Move minijson and safetensor libraries to `examples/utils`. by Quentin Khan · 4 days ago
  27. b407750 Add Pyrefly suppressions by Hana Joo · 4 days ago
  28. 86c3b63 Added run head script to tensor rt rtx end to end by Google AI Edge · 5 days ago
  29. 2c67366 Add LRTModel API to LiteRT Objective-C bindings by Terry Heo · 5 days ago
  30. 14e9107 Add CreateLRTError helper for LiteRT Objective-C APIs by Terry Heo · 5 days ago
  31. 8c98ea8 Internal change by Terry Heo · 5 days ago
  32. a39485f Reverts 72af602586324220f4ba7c0f1e2580fc22aa6b5a by Tenghui Zhu · 5 days ago
  33. 6b7bfd2 Add Gemma4 example to LiteRT copybara export. by Quentin Khan · 5 days ago
  34. f61ee9d Enable running SDPA transposed kernel tests on Mac and OpenCL. by Marissa Ikonomidis · 5 days ago
  35. f5425ab Clean up tests compilation for OSS. by Quentin Khan · 5 days ago
  36. d2f4a1b Centralize Android API and NDK level configurations across LiteRT by Terry Heo · 5 days ago
  37. 3b7d421 Export XNNPACK transformer ops. by Quentin Khan · 5 days ago
  38. b66f705 Export XNNPACK backend and runner. by Quentin Khan · 5 days ago
  39. 7614455 Internal change only. by Google AI Edge · 5 days ago
  40. 9e377c6 Merge pull request #9075 from gunes-arm:arm_compiled_model_api_skeleton_pr by Copybara-Service · 6 days ago
  41. 81e22ed Create an `initialization` shim to handle differences between internal and OSS requirements. by Quentin Khan · 6 days ago
  42. 72af602 Enable IrModel by default in LiteRT by Grant Jensen · 6 days ago
  43. 4d8fa6c Make it possible to use perfetto in OSS. by Quentin Khan · 6 days ago
  44. c44523f [IrModel] Support external shared quantized weights and configure quantized FC by Grant Jensen · 6 days ago
  45. 5867ce6 Harden LiteRT JNI asset model loading in nativeCreateFromAsset by Chunlei Niu · 6 days ago
  46. 4fa7284 Fix tensor buffer lock leaks in LiteRT CompiledModel for CPU tensors. by Andrew Zhang · 6 days ago
  47. 3b748c5 [IrModel] Fix associative op input swapping and Composite KV-Cache parsers by Grant Jensen · 6 days ago
  48. 66a9e10 [IrModel] Extend OneHot and FullyConnected support checks by Grant Jensen · 6 days ago
  49. 0aba14a Add C API interoperability helpers to LiteRt Environment, for internal use only. by Chunlei Niu · 6 days ago
  50. c18781e Fix unsigned quantization type serialization in LiteRT model_utils Quant dialect. by Google AI Edge · 6 days ago
  51. 094a560 Fix IWYU missing absl_log include in gpu_backend_metal by Juhyun Lee · 6 days ago
  52. c3f1d24 Fix buffer overflow in LiteRT Swift API CompiledModel::createOutputBuffers(). by Jun Jiang · 6 days ago
  53. 772f527 Reverts ddf94eb1d43cd9db327d31ce2917eb7649e553c8 by Tenghui Zhu · 6 days ago
  54. ba9695a Improve error message in kernel converter by Google AI Edge · 6 days ago
  55. a600107 Fix type confusion in TFLite resource management where ResourceVariable and StaticHashtable objects shared a common map and were accessed without type checking. CreateResourceVariableIfNotAvailable and CreateHashtableResourceIfNotAvailable now verify existing ResourceType and return a TfLiteStatus. by Adrian Ciausu · 6 days ago
  56. ab08c49 Support optional attention mask for SDPA transposed composite. by Marissa Ikonomidis · 6 days ago
  57. 522683f Support sub-byte types in LiteRT ReferenceTranspose and Transpose generator by Dillon Sharlet · 6 days ago
  58. 104ab19 Add execution tests validating SdpaTransposed via WebGPU param bounds. by Marissa Ikonomidis · 6 days ago
  59. 6881523 Changed gemma3 rms_norm to use transformer_ops. by Misha Gutman · 6 days ago
  60. d4d1ea3 [TFLite] Enforce rank <= 8 with runtime checks in broadcast loop and select. by Dmitri Latushko · 7 days ago
  61. 37fc6b4 Internal change by Terry Heo · 7 days ago
  62. 79501d7 Define and apply TFLITE_PYTHON_CLIENTS_VISIBILITY for Python interpreter by Terry Heo · 7 days ago
  63. ddf94eb When an pipeline stops and restarts an audio stream, destroying and recreating the LiteRT environment encountered exceptions during buffer allocation. by Google AI Edge · 7 days ago
  64. 61de419 Includes names in LiteRT lookup error messages. by David Massoud · 7 days ago
  65. f1f9b40 Internal changes only. by Fengwu Yao · 7 days ago
  66. 249f851 Add tfl quantize op support in the LiteRT model_utils. by Google AI Edge · 7 days ago
  67. 8d72f84 Make LiteRT Swift support Static framework. by Jun Jiang · 7 days ago
  68. 0ec7efc Remove overlooked merge conflict resolution by Gunes Bayir · 7 days ago
  69. 0b51fcf Fix build failure in litert_options by Gunes Bayir · 7 days ago
  70. cea51d0 Add build tests and fix ArmOptions class structure by Gunes Bayir · 7 days ago
  71. 9989064 Skeleton of Arm(R) Compiler Plugin & Dispatch API by Gunes Bayir · 7 days ago
  72. e30325a Internal Change: cleanup by Google AI Edge · 7 days ago
  73. 2dd9ec8 Internal change by Google AI Edge · 8 days ago
  74. 56f2600 Merge pull request #9158 from akodanka:NPU_WS by Copybara-Service · 8 days ago
  75. 0d75062 Quantize user data if model takes quantized inputs. by Andrew Zhang · 8 days ago
  76. e1502e0 Remove the dynamic_runtime directory and targets from LiteRT. by Chunlei Niu · 8 days ago
  77. 11c74eb Merge pull request #9213 from graham0824:dev/chunhsue/int64_bias_tconv2d by Copybara-Service · 8 days ago
  78. cc69924 Update documentation for GPU constant tensor sharing and serialization by Terry Heo · 8 days ago
  79. 96f5cd3 Drop litert/runtime/dispatch:dispatch_delegate_async_test dependency on dynamic_runtime. by Chunlei Niu · 8 days ago
  80. 148eebe Drop dependencies on LiteRT dynamic_runtime targets for Intel OpenVINO compiler targets. by Chunlei Niu · 8 days ago
  81. 309db2b Add perfetto tracing to the end-to-end Gemma 4 example. by Quentin Khan · 8 days ago
  82. 7b7f600 Internal change by Google AI Edge · 8 days ago
  83. f06df48 Google Tensor Compiler Adapter: Update package visibility. by Sarvesh Das · 8 days ago
  84. 72f9a4e Fix non-CPU delegation check in CompiledModel. by Gerardo Carranza · 9 days ago
  85. 004703b Fixing docs.rs documentation generation by Google AI Edge · 10 days ago
  86. cae28e4 Track per-accelerator delegation counts and partition counts in LiteRT and TFLite benchmarker. by Gerardo Carranza · 10 days ago
  87. 6cd19f0 Track per-accelerator delegation counts and partition counts in LiteRT and TFLite benchmarker. by Gerardo Carranza · 10 days ago
  88. a67f269 Mark Conv3D op as supported in Google Tensor compiler plugin. by Google AI Edge · 10 days ago
  89. 35cd6ad Refactor LiteRT Compiler Builder and Matcher C++ APIs by Terry Heo · 11 days ago
  90. afd9894 Refactor LiteRT Options C++ API to use ConcreteOptionsBase by Terry Heo · 11 days ago
  91. aef9d53 Merge pull request #8653 from graham0824:dev/hungjuiw/gemma4-keep-qnodes by Copybara-Service · 11 days ago
  92. ee5f361 Update LiteRT prebuilt binaries and LFS pointers by Google AI Edge · 11 days ago
  93. f49ba44 in ter nal change by David Massoud · 11 days ago
  94. 7f43b53 Add missing <cmath> includes. by Jordan Rupprecht · 11 days ago
  95. 68627c1 - updated CC API to version 2.2.0 by Google AI Edge · 11 days ago
  96. 34342a4 Remove the legacy PIE FullyConnected kernel. by Robert David · 11 days ago
  97. da7f993 Update prebuilt binaries of LiteRT by Google AI Edge · 11 days ago
  98. 19a1585 Bump LiteRT SDK versions to 2.3.0. by Google AI Edge · 11 days ago
  99. 41c72a5 Added detailed comments for dispatch_opaque_options.h by Google AI Edge · 11 days ago
  100. da2c333 Merge pull request #7881 from graham0824:dev/chunhsue/lpai_backend by Copybara-Service · 11 days ago chromium/8015 chromium/8016 chromium/8017 chromium/8018 chromium/8019 chromium/8020 chromium/8021 chromium/8022 chromium/8023 chromium/8024