Skip to content

Android JNI llama cache temperature in class #10287

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Apr 22, 2025
Merged

Android JNI llama cache temperature in class #10287

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c11f8dc
Android JNI llama cache temperature in class
kirklandsign Apr 17, 2025
e8b0a67
Merge branch 'main' into android-temperature-api
kirklandsign Apr 21, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 4 additions & 5 deletions extension/android/jni/jni_layer_llama.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -120,6 +120,7 @@ class ExecuTorchLlmCallbackJni
class ExecuTorchLlmJni : public facebook::jni::HybridClass<ExecuTorchLlmJni> {
private:
friend HybridBase;
float temperature_;
int model_type_category_;
std::unique_ptr<llm::IRunner> runner_;
std::unique_ptr<llm::MultimodalRunner> multi_modal_runner_;
Expand Down Expand Up @@ -175,20 +176,17 @@ class ExecuTorchLlmJni : public facebook::jni::HybridClass<ExecuTorchLlmJni> {
runner_ = std::make_unique<example::Runner>(
model_path->toStdString().c_str(),
tokenizer_path->toStdString().c_str(),
temperature,
data_path->toStdString().c_str());
} else {
runner_ = std::make_unique<example::Runner>(
model_path->toStdString().c_str(),
tokenizer_path->toStdString().c_str(),
temperature);
tokenizer_path->toStdString().c_str());
}
#if defined(EXECUTORCH_BUILD_MEDIATEK)
} else if (model_type_category == MODEL_TYPE_MEDIATEK_LLAMA) {
runner_ = std::make_unique<MTKLlamaRunner>(
model_path->toStdString().c_str(),
tokenizer_path->toStdString().c_str(),
temperature);
tokenizer_path->toStdString().c_str());
// Interpret the model type as LLM
model_type_category_ = MODEL_TYPE_CATEGORY_LLM;
#endif
Expand Down Expand Up @@ -228,6 +226,7 @@ class ExecuTorchLlmJni : public facebook::jni::HybridClass<ExecuTorchLlmJni> {
executorch::extension::llm::GenerationConfig config{
.echo = static_cast<bool>(echo),
.seq_len = seq_len,
.temperature = temperature_,
};
runner_->generate(
prompt->toStdString(),
Expand Down
Loading