docs: add Android LLM runner page and HuggingFace

[omkar-334]

Summary

1. New docs/source/llm/run-on-android.md, a Java reference for the executorch-android AAR runner. Same shape as run-on-ios.md. Covers LlmModule, the LlmModuleConfig builder, LlmGenerationConfig, the LlmCallback methods, load/stop/resetContext, and the image/audio prefill variants. Points at LlamaDemo.

2. Added run-on-android to the LLM toctree in working-with-llms.md, sitting between the Qualcomm page and iOS.

3. In getting-started.md, swapped the two GitHub example links for the in-docs Android and iOS pages so users stay in the docs.

4. Added a tip admonition to using-executorch-export.md under Model Preparation, sending HF Hub users to export-llm-optimum.md before the manual flow.

5. Cleaned up export-llm-optimum.md. Removed the leftover "Method 1" framing since only the CLI path is documented, bumped the orphaned subheadings up a level, and pointed the Running on Device links at the new Android page and the existing iOS page (sample apps kept inline).

Fixes #8790

cc @mergennachin @AlannaBurke @larryliu0820 @cccclai @helunwencser @jackzhxng @byjlw

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/19611

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• Run pull request jobs on OSDC runners in shadow mode

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[github-actions]

This PR needs a release notes: label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[Copilot]

Pull request overview

Adds first-party documentation for running LLMs on Android via the executorch-android AAR and improves the Hugging Face (Optimum ExecuTorch) discovery flow across the LLM docs.

Changes:

• Added a new Android LLM runner guide (run-on-android.md) documenting LlmModule, LlmModuleConfig, LlmGenerationConfig, callbacks, and multimodal prefill APIs.
• Updated LLM docs navigation to include the Android guide and adjusted “Getting Started” running links to point to in-doc pages.
• Improved Hugging Face export guidance by adding an Optimum tip in the export docs and cleaning up the Optimum export page (headings + device-running links).

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
docs/source/using-executorch-export.md	Adds a tip directing Hugging Face Hub users to the Optimum ExecuTorch export flow.
docs/source/llm/working-with-llms.md	Inserts run-on-android into the LLM toctree.
docs/source/llm/run-on-android.md	New Android Java runtime guide for LlmModule + configs + multimodal prefill APIs.
docs/source/llm/getting-started.md	Updates “Running” links to point to in-doc Android/iOS pages.
docs/source/llm/export-llm-optimum.md	Renames/restructures CLI export section and updates “Running on device” links to new docs pages.

Comments suppressed due to low confidence (2)

docs/source/llm/run-on-android.md:113

• LlmGenerationConfig docs list maxNewTokens and warming as supported generation parameters, but LlmModule.generate(String, LlmGenerationConfig, ...) currently only reads seqLen, echo, temperature, numBos, and numEos (it never uses maxNewTokens / warming). This makes the documentation misleading because setting those fields has no effect. Consider either updating the doc to call out which fields are currently honored, or updating the Java binding/native call to plumb maxNewTokens/warming through if supported by the underlying runner.

For full control over generation parameters, use `LlmGenerationConfig`:

```java
LlmGenerationConfig genConfig = LlmGenerationConfig.create()
    .seqLen(2048)
    .temperature(0.8f)
    .echo(false)
    .build();

module.generate("Once upon a time", genConfig, callback);

LlmGenerationConfig exposes echo, maxNewTokens, seqLen, temperature, numBos, numEos, and warming. Defaults match the C++ GenerationConfig documented in Running LLMs with C++.

**docs/source/llm/run-on-android.md:164**
* In the normalized-image `ByteBuffer` example, after writing floats into `floatBuffer` the buffer position will typically be at the end, so calling `prefillNormalizedImage(floatBuffer, ...)` will fail validation due to insufficient `remaining()` bytes. The example should reset the position (e.g., `flip()`/`rewind()`) after filling the buffer, similar to the raw-byte example above.

ByteBuffer floatBuffer = ByteBuffer
.allocateDirect(3 * 336 * 336 * Float.BYTES)
.order(ByteOrder.nativeOrder());
// fill floatBuffer with normalized values, then:
module.prefillNormalizedImage(floatBuffer, 336, 336, 3);

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.