Skip to content

Internal only change.#1560

Closed
copybara-service[bot] wants to merge 1157 commits intomainfrom
litert_lm_pr_878801390
Closed

Internal only change.#1560
copybara-service[bot] wants to merge 1157 commits intomainfrom
litert_lm_pr_878801390

Conversation

@copybara-service
Copy link
Copy Markdown
Contributor

Internal only change.

yishuangP and others added 30 commits January 14, 2026 10:34
LiteRT-LM-PiperOrigin-RevId: 856271008
LiteRT-LM-PiperOrigin-RevId: 856697433
**Problems**:
1. engine selection logic is duplicated in multiple places.
2. not all targets use the selection logic. The selection flag works only for some targets, confusing developers.
3. This leads to issues like duplicated symbols (CL/856236129) and building the unexpected engine.

**Changes**
- Create a target ":engine_impl_selected" to select the implementation
- Migrate selection logic and hardcoded implementation to ":engine_impl_selected"
- limit the visibility of existing engine implementations

LiteRT-LM-PiperOrigin-RevId: 856697765
LiteRT-LM-PiperOrigin-RevId: 856733960
LiteRT-LM-PiperOrigin-RevId: 856790867
**New capabilities**:
1. allow creating conversation with initial messages.
2. new simplified sendMessage and sendMessageAsync.

**Changes**:
- Define enum Role {SYSTEM, USER, MODEL}
- Add Role to "Message" class.
- Add `initialMessages` to `ConversationConfig` to set the init messages
- Add new sendMessage*() with text and list of contents
- Deprecate the systemMessage since it is covered by initialMessages.
- Deprecate Message.of factory methods. Favor the new system/user/model factory methods.

LiteRT-LM-PiperOrigin-RevId: 857227983
1. mention FunctionGemma
2. update sendMessage signatures missed in previous change.

LiteRT-LM-PiperOrigin-RevId: 857262567
LiteRT-LM-PiperOrigin-RevId: 857335054
LiteRT-LM-PiperOrigin-RevId: 857350704
LiteRT-LM-PiperOrigin-RevId: 857357241
LiteRT-LM-PiperOrigin-RevId: 857423226
Follow the set_* pattern to make it scalable instead of adding all value to the single C function.

LiteRT-LM-PiperOrigin-RevId: 857689534
LiteRT-LM-PiperOrigin-RevId: 858607999
LiteRT-LM-PiperOrigin-RevId: 858656280
LiteRT-LM-PiperOrigin-RevId: 858718908
LiteRT-LM-PiperOrigin-RevId: 858742888
LiteRT-LM-PiperOrigin-RevId: 858748258
LiteRT-LM-PiperOrigin-RevId: 858802033
LiteRT-LM-PiperOrigin-RevId: 858833664
LiteRT-LM-PiperOrigin-RevId: 859090887
Jetson doesn't have std::logf().
Use std::log(static_cast<float>()) instead.

LiteRT-LM-PiperOrigin-RevId: 859148505
- Transpose mask in input handling for fixed gpu kernel size
- Fix exported symbols from libLiteRt.so on linux_arm64

LiteRT-LM-PiperOrigin-RevId: 859164020
LiteRT-LM-PiperOrigin-RevId: 859247355
Currently, tools could be implemented with Kotlin functions only. It is easy to use but does not give developer the full control over the tools spec.

For users who need fine-grained control over the tool description and execution, they can use the new `OpenApiTool` now.

For examples how to implement it, see the `README.md`.

LiteRT-LM-PiperOrigin-RevId: 859271963
- Last sampler shlibs have some regressions

LiteRT-LM-PiperOrigin-RevId: 859350407
LiteRT-LM-PiperOrigin-RevId: 859642028
When context length is big, it increase init time noticeably, e.g. 8s when context length = 32k for gemma3-1b.

LiteRT-LM-PiperOrigin-RevId: 859666933
LiteRT-LM-PiperOrigin-RevId: 859859182
hheydary and others added 22 commits March 2, 2026 08:37
LiteRT-LM-PiperOrigin-RevId: 877415471
Add x86_64-linux-android to supported platforms in WORKSPACE.
Update cargo-bazel-lock.json to reflect the platform changes and dependency resolution.

LiteRT-LM-PiperOrigin-RevId: 877443693
LiteRT-LM-PiperOrigin-RevId: 877456830
LiteRT-LM-PiperOrigin-RevId: 877463951
LiteRT-LM-PiperOrigin-RevId: 877486313
LiteRT-LM-PiperOrigin-RevId: 877536707
Implement patchify for stb image processor
Support map of TensorBuffers in InputImage.

LiteRT-LM-PiperOrigin-RevId: 877682820
LiteRT-LM-PiperOrigin-RevId: 878014141
LiteRT-LM-PiperOrigin-RevId: 878052506
LiteRT-LM-PiperOrigin-RevId: 878095579
LiteRT-LM-PiperOrigin-RevId: 878143704
Support end of vision tflite model in executors

LiteRT-LM-PiperOrigin-RevId: 878252087
Integrate model downloading into macOS, Windows, and Linux CI workflows using a Gemma 3 1B IT model from Hugging Face. Introduce a cross-platform Pytest framework to execute litert_lm_main with a single prompt, serving as an E2E smoke test to verify basic inference functionality and expected output.

LiteRT-LM-PiperOrigin-RevId: 878581563
LiteRT-LM-PiperOrigin-RevId: 878609747
LiteRT-LM-PiperOrigin-RevId: 878717839
LiteRT-LM-PiperOrigin-RevId: 878885676
LiteRT-LM-PiperOrigin-RevId: 879067681
LiteRT-LM-PiperOrigin-RevId: 879202010
@copybara-service copybara-service Bot force-pushed the litert_lm_pr_878801390 branch from 25e82cb to a395274 Compare March 5, 2026 21:31
LiteRT-LM-PiperOrigin-RevId: 879221034
@copybara-service copybara-service Bot force-pushed the litert_lm_pr_878801390 branch from a395274 to 7abedd3 Compare March 5, 2026 21:32
ai-edge-bot and others added 2 commits March 5, 2026 15:19
LiteRT-LM-PiperOrigin-RevId: 879268560
LiteRT-LM-PiperOrigin-RevId: 878801390
@copybara-service copybara-service Bot force-pushed the litert_lm_pr_878801390 branch from 7abedd3 to feb4d9d Compare March 5, 2026 23:22
@TomOtero1984 TomOtero1984 force-pushed the litert_lm_pr_878801390 branch from feb4d9d to 06ecae9 Compare March 9, 2026 21:12
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.