Arm Examples
Official@arm-examples
Offers specialized engineering procedures for cross-compiling, benchmarking, and validating high-performance inference libraries on AArch64 and Android architectures.
Agent Skills by Arm Examples
Showing 12 vetted skills indexed across 1 GitHub repositories.
Arm-Examplesllm-backend-scaffold
Scaffolds new LLM backends into the LLM-Runner repository.
Arm-Examplesllm-debug-test-failures
Debug failing LLM integration tests with verbose ctest reruns and Python scripts.
Arm-Examplesllm-code-review
Review code changes for correctness, tests, documentation, portability, and performance.
Arm-Examplesllm-session-start
Run environment and wiring sanity checks for the LLM repository.
Arm-Examplesllm-jni-smoke
Validate the Java Native Interface build and test environment for LLM applications.
Arm-Examplesllm-add-model-support
Add and revise LLM model configuration JSON files with CTest integration.
Arm-Examplesllm-benchmark-workflow
Build and execute LLM benchmarking tools across multiple backends.
Arm-Examplesllm-change-api-safely
Modify the public C++ API and JNI implementations with synchronized updates and testing.
Arm-Examplesllm-build-and-ctest
Configure, build, and run CTest for the LLM library using CMake presets.
Arm-Examplesllm-android-cross-build
Cross-compile LLM libraries for Android AArch64 using CMake presets and the Android NDK.
Arm-Examplesllm-update-downloads
Update URLs and SHA256 checksums in requirements.json for LLM model downloads.
Arm-Examplesllm-config-schema-change
Modify LLM configuration schemas by updating parsing logic, tests, and documentation for JSON config keys.
Frequently Asked Questions About Arm Examples
FAQPage SchemaWhat specific tasks are enabled by these engineering procedures?▼
These procedures enable cross-compilation for Android AArch64, validation of Java Native Interface builds, execution of performance benchmarks, and management of configuration schemas. They facilitate the integration of new model architectures into existing C++ inference backends while ensuring build integrity through standardized testing.
Which target personas benefit from these technical procedures?▼
These procedures are designed for embedded systems engineers, performance optimization specialists, and mobile developers working on AArch64 hardware. They are specifically intended for developers maintaining high-performance inference libraries who require consistent build, test, and deployment cycles across diverse hardware backends.
What are the prerequisites for executing these build and test procedures?▼
Execution requires a configured environment with CMake, the Android NDK for cross-compilation, and a local clone of the target repository. Users must ensure their environment supports CTest for validation and that all necessary dependencies defined in the requirements JSON files are present.