Introduce LoraConfig by lucylq · Pull Request #17229 · pytorch/executorch

lucylq · 2026-02-05T02:22:10Z

Stack from ghstack (oldest at bottom):

Differential Revision: D92304723

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

pytorch-bot · 2026-02-05T02:22:14Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/17229

📄 Preview Python docs built from this PR

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

❌ 4 New Failures, 6 Unrelated Failures

As of commit 53a5176 with merge base aa2f683 ():

NEW FAILURES - The following jobs have failed:

Test CUDA Builds / check-all-cuda-builds (gh)
Process completed with exit code 1.
Test CUDA Builds / test-executorch-cuda-build-12.6 / linux-job (gh)
RuntimeError: Command docker exec -t 58d02383d136108bc8a5b3182a0a0e48fd658de0e3cbc844398b188bc713fa02 /exec failed with exit code 1
trunk / test-qnn-optimum-model (fp32, efficientnet) / linux-job (gh)
RuntimeError: Command docker exec -t 0b6823792c2e20a94e7a0b9e15f2f7469edcfb9df3d28b7400920f1cf80e039c /exec failed with exit code 92
trunk / test-static-llama-qnn-eval-linux (baseline, 62) / linux-job (gh)
RuntimeError: Command docker exec -t dc6b8efc65b0a3c957f87545fa5b2a79078dd1327bdd4c9bd00cef07dd9dde5f /exec failed with exit code 1

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

Test CUDA Builds / export-model-cuda-artifact (mistralai, Voxtral-Mini-3B-2507, non-quantized) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)
Test CUDA Builds / export-model-cuda-artifact (nvidia, parakeet-tdt, quantized-int4-tile-packed) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)
Test CUDA Builds / export-model-cuda-artifact (openai, whisper-large-v3-turbo, quantized-int4-weight-only) / linux-job (gh) (matched linux rule in flaky-rules.json)
The process '/usr/bin/git' failed with exit code 128
Test CUDA Builds / test-executorch-cuda-build-12.9 / linux-job (gh) (detected as infra flaky with no log or failing log classifier)
Test CUDA Builds / test-models-cuda (add_mul) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)
Test CUDA Builds / test-models-cuda (mv3) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)

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

github-actions · 2026-02-05T02:23:11Z

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.

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

Copilot

Pull request overview

This PR introduces a LoraConfig dataclass to encapsulate LoRA (Low-Rank Adaptation) adapter configuration, replacing the previous separate adapter_checkpoint and adapter_config fields in BaseConfig. This refactoring improves code organization and provides a cleaner API for LoRA configuration.

Changes:

Added LoraConfig dataclass with validation and automatic config parsing from JSON
Refactored BaseConfig to use LoraConfig instead of separate adapter fields
Updated from_args method to create LoraConfig instances from CLI arguments
Simplified adapter loading logic in Llama2Model by using the new LoraConfig object

Reviewed changes

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

File	Description
extension/llm/export/config/llm_config.py	Introduces `LoraConfig` dataclass with validation, updates `BaseConfig` to use it, and modifies `from_args` to create `LoraConfig` from CLI arguments
examples/models/llama/model.py	Refactors adapter loading to use `LoraConfig` object, simplifying the logic by removing redundant validation and JSON parsing

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

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

@hakanb

Pull Request resolved: #17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340400147 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/)

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

@hakanb

Pull Request resolved: #17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340423626 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/)

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

@hakanb

Pull Request resolved: #17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340434171 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/)

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

@hakanb

Pull Request resolved: #17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340443733 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/)

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

@hakanb

Pull Request resolved: #17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340443733 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/)

kimishpatel

Review automatically exported from Phabricator review in Meta.

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

@hakanb

Pull Request resolved: #17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340525244 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/)

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

@hakanb

Pull Request resolved: #17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340525244 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/)

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

@hakanb

Pull Request resolved: #17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340525244 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/)

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

@hakanb

Pull Request resolved: #17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340525244 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/)

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

@hakanb

Pull Request resolved: #17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340930093 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/)

@lucylq

This PR was created by the merge bot to help merge the original PR into the main branch. ghstack PR number: #17229 by @lucylq ^ Please use this as the source of truth for the PR details, comments, and reviews ghstack PR base: https://github.com/pytorch/executorch/tree/gh/lucylq/132/base ghstack PR head: https://github.com/pytorch/executorch/tree/gh/lucylq/132/head Merge bot PR base: https://github.com/pytorch/executorch/tree/gh/lucylq/131/orig Merge bot PR head: https://github.com/pytorch/executorch/tree/gh/lucylq/132/orig Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) @diff-train-skip-merge Co-authored-by: Github Executorch <github_executorch@arm.com>

@hakanb

Pull Request resolved: #17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340930093 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/)

@hakanb

Pull Request resolved: #17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340930093 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) Co-authored-by: Github Executorch <github_executorch@arm.com>

@hakanb

Pull Request resolved: pytorch#17229 Introduce LoraConfig to hold lora parameters such as: - checkpoint - rank - target_modules (e.g. q_proj, k_proj, v_proj, up_proj, down_proj, gate_proj, o_proj) - lora_alpha LoraConfig validation done post-init. LoraConfig can be created with config.json file. Update cases of export_llama_lib to use LoraConfig instead of adapter_checkpoint and adapter_config. NOTE: we may need to extend this to support more customizable features like lora config per layer etc. cc @hakanb ghstack-source-id: 340930093 @exported-using-ghexport Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) Co-authored-by: Github Executorch <github_executorch@arm.com>

Introduce LoraConfig

70faa5b

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

lucylq requested review from larryliu0820 and mergennachin as code owners February 5, 2026 02:22

This was referenced Feb 5, 2026

Support multimethod in runner #17228

Merged

Introduce MultimethodLoraConfig #17230

Merged

Support multimethod in export_llama_lib #17231

Merged

meta-cla Bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Feb 5, 2026

meta-codesync Bot added fb-exported meta-exported labels Feb 5, 2026

lucylq mentioned this pull request Feb 5, 2026

Support multimethod in export_llama #17010

Closed

lucylq changed the title ~~Introduce LoraConfig~~ [wip] Introduce LoraConfig Feb 5, 2026

Update on "[wip] Introduce LoraConfig"

9876e7d

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

lucylq changed the title ~~[wip] Introduce LoraConfig~~ Introduce LoraConfig Feb 5, 2026

lucylq requested a review from Copilot February 5, 2026 19:01

Copilot started reviewing on behalf of lucylq February 5, 2026 19:02 View session

Copilot AI reviewed Feb 5, 2026

View reviewed changes

Comment thread extension/llm/export/config/llm_config.py Outdated

Comment thread extension/llm/export/config/llm_config.py Outdated

Comment thread extension/llm/export/config/llm_config.py Outdated

Github Executorch added 2 commits February 10, 2026 08:59

Update on "Introduce LoraConfig"

e4e9f42

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

Update on "Introduce LoraConfig"

e3022d2

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

Update on "Introduce LoraConfig"

812286c

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

Update on "Introduce LoraConfig"

3e5c5d9

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

Update on "Introduce LoraConfig"

e73d873

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

Update on "Introduce LoraConfig"

f696cf1

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

kimishpatel approved these changes Feb 12, 2026

View reviewed changes

Update on "Introduce LoraConfig"

12c2bc8

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

Update on "Introduce LoraConfig"

7124c01

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

Update on "Introduce LoraConfig"

deaa1b4

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

Update on "Introduce LoraConfig"

22843be

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

Update on "Introduce LoraConfig"

53a5176

Differential Revision: [D92304723](https://our.internmc.facebook.com/intern/diff/D92304723/) [ghstack-poisoned]

meta-codesync Bot merged commit 3ac1f31 into gh/lucylq/132/base Feb 13, 2026
314 of 326 checks passed

meta-codesync Bot deleted the gh/lucylq/132/head branch February 13, 2026 07:01

meta-codesync Bot temporarily deployed to cherry-pick-bot February 13, 2026 07:01 Inactive

pytorchbot mentioned this pull request Feb 13, 2026

Introduce LoraConfig #17436

Merged

lucylq mentioned this pull request Feb 13, 2026

Introduce LoraConfig #17458

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Introduce LoraConfig#17229

Introduce LoraConfig#17229
meta-codesync[bot] merged 13 commits intogh/lucylq/132/basefrom
gh/lucylq/132/head

lucylq commented Feb 5, 2026 •

edited

Loading

Uh oh!

pytorch-bot Bot commented Feb 5, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Feb 5, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

kimishpatel left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

lucylq commented Feb 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot Bot commented Feb 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/17229

❌ 4 New Failures, 6 Unrelated Failures

Uh oh!

github-actions Bot commented Feb 5, 2026

This PR needs a release notes: label

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

kimishpatel left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

lucylq commented Feb 5, 2026 •

edited

Loading

pytorch-bot Bot commented Feb 5, 2026 •

edited

Loading

This PR needs a `release notes:` label