Google GitGoogle Git
Sign in
chromium/external/github.com/google-ai-edge/LiteRT-LM/HEAD
  1. e06a29f Add support for streaming HF_Tokenizer_Zlib sections in LiteRT-LM by Google AI Edge · 5 hours ago upstream/main
  2. c05e67a Update dependencies of litert_lm by Changming Sun CA · 5 hours ago
  3. df97cc1 Close the Kokoro TTS front-end gap to Host Kokoro for espeak and CJK languages. by Yi-Chun Kuo · 6 hours ago
  4. 039bd37 When ThinkingBudgetConstraint is configured with budget == 0, it by Google AI Edge · 6 hours ago
  5. 3981a9a Calibrate STFT framing and Hann window for Whisper. by Byungchul Kim · 7 hours ago
  6. cfa635c Support Japanese in Kokoro TTS with bundled MeCab/UniDic lexicon and G2P. by Yi-Chun Kuo · 10 hours ago
  7. b7c40ea Update dependencies of litert_lm by Fengwu Yao · 10 hours ago
  8. cb97cdd Handle silence more reliably. by Byungchul Kim · 11 hours ago
  9. 40d6cda Stage only dated nightly XCFrameworks and checksums by Google AI Edge · 11 hours ago
  10. 2e79135 Migrate remaining NPU option callers to GetOptions<T>() by Terry Heo · 12 hours ago
  11. 4fad493 Migrate ASR metadata to static model/factory registration to OmniEngine. by Byungchul Kim · 13 hours ago
  12. b2f2d9d Align overlapping ASR chunks more correctly in text mergers. by Byungchul Kim · 14 hours ago
  13. a484d21 Handle empty `session_inputs` before calling `RunPrefillAsync` in by Google AI Edge · 25 hours ago
  14. a82895c Automated Code Change by Google AI Edge · 28 hours ago
  15. 9fbce89 Allow for prefix caching for artisan GPU models in LiteRT-LM CLI by Google AI Edge · 29 hours ago
  16. 0ae5873 Map subword tokens to word-onset timestamps in TokenizerDetokenizer. by Byungchul Kim · 30 hours ago
  17. e609517 add bidirectional LSTM custom op in omni base by Yi-Chun Kuo · 30 hours ago
  18. 880b3db Add bridge between legacy C++ ModelType and proto TfLiteModelType enums. by Yi-Chun Kuo · 32 hours ago
  19. b4a2027 Enable Slaney mel filterbanks and power spectrum in ASR model metadata. by Byungchul Kim · 33 hours ago
  20. 9ab8b8d Support multi-token decode_start_token_ids across ASR decoders for Whisper. by Byungchul Kim · 34 hours ago
  21. c10c6c4 Support Swift nightly package distribution and fix macOS codesigning. by Google AI Edge · 34 hours ago
  22. 7659436 Copy host memory state snapshots into device buffers when restoring state. by Mohammadreza Heydary · 34 hours ago
  23. d6272ea Support Slaney mel filterbank and power-spectrum support to MelFilterbank. by Byungchul Kim · 34 hours ago
  24. 200545f Downstream mobile and embedded deployments of LiteRT-LM require fine-grained control over optional tokenizers, template engines, grammar parsers, constrained decoding engines, model-family data processors, and hardware backends to minimize binary size. Previously, ad-hoc `--define` flags were scattered across individual `BUILD` packages. by Salil Tambe · 35 hours ago
  25. b1faae7 Internal change by Mohammadreza Heydary · 2 days ago
  26. 35b09b1 Stream all model loads in the LiteRT-LM Web SDK and default to the GPU backend by Google AI Edge · 2 days ago
  27. 81c758e Refactor `model_data_processor_factory` to use `ModelDataProcessorRegistry`. by Salil Tambe · 2 days ago
  28. b966041 Fix overflow issue in npu embedder by Google AI Edge · 2 days ago
  29. 03d3979 Add antlr_parser_utils_noop target for builds without ANTLR/Rust tool-use parsers. by Salil Tambe · 4 days ago
  30. 98524f6 Decouple LlgConstraintProvider header from Rust llguidance FFI and add noop targets. by Salil Tambe · 4 days ago
  31. 4d8a9a2 Decouple `PromptTemplate` header from Rust `minijinja_template` FFI and add noop target. by Salil Tambe · 4 days ago
  32. a50b5f5 Add streaming model loading support to LiteRT-LM Engine by Google AI Edge · 4 days ago
  33. cb17ef0 Fix macOS/iOS symbol visibility in `liblitert-lm.so` to prevent load-time crashes. by Google AI Edge · 4 days ago
  34. d99b438 Update dependencies of litert_lm by Mohammadreza Heydary · 4 days ago
  35. f3832ed No public description by Google AI Edge · 5 days ago
  36. b0529f4 Configure Git LFS endpoint in .lfsconfig for LiteRT-LM. by Google AI Edge · 5 days ago
  37. 8111f12 [LiteRT-LM] Remove synchronous NPU warmup inference from EmbeddingLiteRtCompiledModelExecutor::CompileTextEncoder by Google AI Edge · 5 days ago
  38. f42a1bb Add C API version macros and litert_lm_get_api_version so consumers can detect header/library version skew. by Tenghui Zhu · 5 days ago
  39. 4a363c0 New session configs based on protobuf by Byungchul Kim · 5 days ago
  40. 798d51e Support FLUX.2-klein engine/session by Byungchul Kim · 5 days ago
  41. 2d872e0 Add Android benchmarks for Gemma 4 2B LoRA on CPU and GPU. by Salil Tambe · 5 days ago
  42. 0063187 Allow creating embedding engines for models without a vision encoder. by Google AI Edge · 5 days ago
  43. b2f3d3c Add speculative decoding statistics to the MTP drafter, the executor and the by Mohammadreza Heydary · 5 days ago
  44. 8b7bb5c When the thought channel is not prefilled and the model skips thinking, by Google AI Edge · 5 days ago
  45. 4697c72 - Write all-ones into `kPrevMaskName` in `input_buffers_map_` and clear `buffered_spectrogram_` inside `AudioLiteRtCompiledModelExecutor::AudioStreamingEncoder::CreateNewContext` so new streaming contexts start cleanly. by Google AI Edge · 6 days ago
  46. e16809a Handle partial chunks correctly in AudioInputSource and TextInputSource. by Byungchul Kim · 6 days ago
  47. ea11963 Document in the C API that sessions and conversations must be destroyed before the engine they were created from. by Tenghui Zhu · 6 days ago
  48. 74e702d Text2Image engine and session by Byungchul Kim · 6 days ago
  49. d07a923 Fix Gemma-4 GPU LoRA support for quantized models in MLDrift delegate. by Salil Tambe · 6 days ago
  50. 4b15b34 Report LoRA inputs that don't match any tensor in the LoRA data. by Google AI Edge · 6 days ago
  51. e9fb7ff Open-source the EmbeddingGemma 2 multimodal search web demo by Google AI Edge · 6 days ago
  52. 97e6dc7 [LiteRT-LM] Eagerly filter channel content from KV cache after decode. by Google AI Edge · 6 days ago
  53. b8b3131 Fix text scoring rewinding the KV-cache position twice after prefill. by Google AI Edge · 6 days ago
  54. 7840057 Isolate espeak_ng genrule working directories to fix parallel build race in standalone Bazel builds by Salil Tambe · 6 days ago
  55. 7233a2d Create LoRA buffers per signature so prefill gets buffer types it supports by Google AI Edge · 6 days ago
  56. 3a1f60d Rename Omni ASR/TTS session targets and standardize stage/factory signatures. by Byungchul Kim · 6 days ago
  57. d4beba4 Fix multibyte UTF-8 handling and byte-fallback detokenization in BufferedStreamingDetokenizer. by Mohammadreza Heydary · 6 days ago
  58. 1f7fa76 Add FLUX.2 model config, math helpers, denoiser stage, and VAE decoder stage. by Byungchul Kim · 6 days ago
  59. cb1495f update benchmark script by Google AI Edge · 7 days ago
  60. 4ecac4f Add DMA-BUF memory metrics to LiteRT LM metrics proto. by Andrew Zhang · 7 days ago
  61. 4c72b51 Unify ASR and TTS session orchestration under MultiStagedSession. by Byungchul Kim · 7 days ago
  62. f0f235e Add embedding demo link to litert lm readme by Google AI Edge · 7 days ago
  63. a926bb0 Fill lora_param_tensor so GPU runtime-BMM LoRA covers all channels by Salil Tambe · 7 days ago
  64. 5b87f5b Add TextEncoderStage under omni/text2image. by Byungchul Kim · 7 days ago
  65. 93d7412 Bump LiteRT-LM version to 0.19.0 by Google AI Edge · 7 days ago
  66. deaaa64 Re-prime the MTP drafter after any non-speculative step. by Mohammadreza Heydary · 8 days ago
  67. 8eca57a Add Conversation::CountTokens() API to measure prospective message token footprint. by Google AI Edge · 8 days ago
  68. c9848c4 add v0.18.0 release notes by Google AI Edge · 8 days ago
  69. 0bb486b Load Gemma 4 audio encoder LoRAs under the encoder's own input names by Salil Tambe · 8 days ago
  70. 84d9669 Pipe through max_top_k_ setting when set by Google AI Edge · 8 days ago
  71. dbdb99b Internal change by Google AI Edge · 8 days ago
  72. 525de7e Add an --output_size flag to embedding_litert_lm_main for truncation of the output embedding. by Marissa Ikonomidis · 8 days ago
  73. 0540e5c Replace `# pytype: disable` suppressions with `# pyrefly: ignore` by Oleh Prypin · 9 days ago
  74. d17a52f Add ImageDecoder and PromptSource interface for image gen model inference by Byungchul Kim · 11 days ago
  75. 10245c8 Internal change by Wai Hon Law · 11 days ago
  76. 8092fe5 Extend the Kotlin `benchmark` API and JNI bindings (`nativeCreateBenchmark`) to accept optional `visionBackend`, `audioBackend`, and `contents` parameters, and expose `BenchmarkInfo.markDurationsInSecond` from `BenchmarkInfo::GetMarkDurations()`. by Salil Tambe · 11 days ago
  77. 5befccd Restrict espeak-ng dictionaries to Kokoro-supported languages. by Tenghui Zhu · 11 days ago
  78. f136f4a Handle silence in more robust way by Byungchul Kim · 11 days ago
  79. 400273c Add pre-compiled model support to LiteRT-LM CompiledModelExecutor by Google AI Edge · 11 days ago
  80. b320801 Internal change by Tenghui Zhu · 12 days ago
  81. 956de4b Strengthen input validation and null pointer checks in LiteRT LM C APIs. by Mohammadreza Heydary · 12 days ago
  82. 58fcc1c Preserve prefill start token and truncate repetition loops in LmDecoder. by Byungchul Kim · 12 days ago
  83. b34761f Stitch overlapping chunks at longest contiguous match in LevenshteinTextMerger. by Byungchul Kim · 12 days ago
  84. c8ff070 Retain unconfirmed words on empty chunks in TimestampTextMerger. by Byungchul Kim · 12 days ago
  85. 3dbb23e Skip auxiliary mask and cache update warmup in LiteRT-LM NPU compiled model executor when using hardware update (kWH). by Andrew Zhang · 12 days ago
  86. 1a14a11 Remove litert_lm_get_last_error_code from the LiteRT-LM C API. by Mohammadreza Heydary · 12 days ago
  87. 5775c07 Zero-pad wrapped ring buffer frames at EOF in FileAudioSource. by Byungchul Kim · 12 days ago
  88. 4af508a Document the LiteRT-LM C API 1.0.0 return-shape contract by Mohammadreza Heydary · 12 days ago
  89. fc6ca8a Return status codes from the experimental LiteRT-LM C API by Mohammadreza Heydary · 12 days ago
  90. a5d53ea Return status codes from all LiteRT-LM model info C API functions.\ by Mohammadreza Heydary · 12 days ago
  91. 216b00b Add ASR, TTS, and ImageGen in litert_lm_builder. by Byungchul Kim · 12 days ago
  92. 5190782 Return status codes from all LiteRT-LM embedding engine C API functions. by Mohammadreza Heydary · 12 days ago
  93. b4a9d0e Return status codes from LiteRT-LM C API conversation functions by Mohammadreza Heydary · 12 days ago
  94. f4455da Sanitize tool and property names for Lark grammar rules in constrained decoding. by Google AI Edge · 12 days ago
  95. 8ab3da0 Return status codes from the remaining LiteRT-LM C engine.h accessors by Mohammadreza Heydary · 12 days ago
  96. 7d9a2a3 Return status codes from LiteRT-LM C API result-producing functions by Mohammadreza Heydary · 13 days ago
  97. 1dc4633 Return status codes from LiteRT-LM C API engine constructors by Mohammadreza Heydary · 13 days ago
  98. d101ec9 Make the dummy test model's output shape match its declared shape. by Tommy Chiang · 13 days ago
  99. 9369b17 Fix support for Gemma3 270M in LiteRT LM runtime and builder. by Andrew Zhang · 13 days ago
  100. 12ba057 Also skip the host local attention mask fill when both masks are pruned by the GPU delegate. by Fengwu Yao · 13 days ago