Make multimethod generic in llm_config

[lucylq]

Stack from ghstack (oldest at bottom):

• -> Make multimethod generic in llm_config #18213

Differential Revision: D96822523

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/18213

• 📄 Preview Python docs built from this PR

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

✅ You can merge normally! (2 Unrelated Failures)

As of commit f42c141 with merge base 28b4813 ():

BROKEN TRUNK - The following jobs failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

• pull / unittest / windows / windows-job (gh) (trunk failure)
  ##[error]The operation was canceled.
• pull / unittest-editable / windows / windows-job (gh) (trunk failure)
  ##[error]The operation was canceled.

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.

[kimishpatel]

Review automatically exported from Phabricator review in Meta.

[Copilot]

Pull request overview

This PR generalizes the LLM export “multimethod” configuration by renaming the previous multimethod_lora configuration to a more generic multimethod and switching from a dict-based schema to a list of per-method config objects, then updating the example YAML and llama export path accordingly.

Changes:

• Replace MultimethodLoraConfig with MultimethodConfig + new MethodConfig (list-based schema) in LlmConfig.
• Update Qwen3 multimethod example YAML to the new schema.
• Update llama export/validation logic to consume llm_config.multimethod and iterate through method configs.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File	Description
extension/llm/export/config/llm_config.py	Introduces MethodConfig and replaces multimethod_lora with a generic multimethod list-based configuration.
examples/models/qwen3/config/qwen3_multimethod.yaml	Updates the Qwen3 multimethod example to use the new multimethod.methods list format.
examples/models/llama/export_llama_lib.py	Updates validation and multimethod export logic to use the new config shape and naming.

---

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

[Copilot]

The MultimethodConfig docstring still says it "Maps method names" even though methods is now a List[MethodConfig]. This is misleading for config authors; please update the docstring wording to reflect the list-based schema (and, if relevant, that each list element’s lora_config=None means base weights).

[Copilot]

With multimethod.methods now being a list, duplicate method_name entries are possible. The current code will silently overwrite earlier entries in method_to_program[...], but num_methods/logging still counts duplicates, which can lead to confusing or incorrect exports. Please validate that all method.method_name values are unique (and non-empty) before populating method_to_program, and raise a clear error if duplicates are found.