[Refactor] Separate `_prepare_inputs` to `_prepare_inputs` and `_preprocess` by gcanlin · Pull Request #6191 · vllm-project/vllm-ascend

gcanlin · 2026-01-23T08:12:37Z

What this PR does / why we need it?

Align with upstream vLLM. This PR will help downstream vLLM-Omni reduce the cost for maintaining the _prepare_inputs. Besides, it helps vLLM-Ascend code more readable. In the future, we can follow closer to vLLM. The preprocess logic is same as GPUModelRunner. We don't need to maintain it anymore.

Does this PR introduce any user-facing change?

No

How was this patch tested?

CI.

vLLM version: v0.14.0
vLLM main: vllm-project/vllm@d682094

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

github-actions · 2026-01-23T08:13:36Z

👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:‌‌

A PR should do only one thing, smaller PRs enable faster reviews.
Every PR should include unit tests and end-to-end tests ‌to ensure it works and is not broken by other future PRs.
Write the commit message by fulfilling the PR description to help reviewer and future developers understand.

If CI fails, you can run linting and testing checks locally according Contributing and Testing.

gemini-code-assist

Code Review

This pull request refactors the _prepare_inputs method by splitting it into _prepare_inputs and _preprocess, aligning it with the upstream vLLM implementation. This is a good step towards improving code readability and maintainability.

My review focuses on ensuring the refactoring is clean and doesn't introduce new issues. I've identified a couple of minor cleanup opportunities: an unused parameter in the refactored _prepare_inputs method and a confusing TODO comment that appears to be a typo. Addressing these will further improve the code quality.

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gcanlin · 2026-01-23T14:07:20Z

@wangxiyuan CI has passed.
@zhenwenqi2024 @kunpengW-code If #6043 can't be ready, we will merge this first. It's better that you can also take a look at this PR.

github-actions · 2026-01-24T12:14:38Z

This pull request has conflicts, please resolve those before we can evaluate the pull request.

wangxiyuan · 2026-01-25T04:02:47Z

CI failed

wangxiyuan · 2026-01-25T04:14:41Z

I think it's related to #6041

gcanlin · 2026-01-25T04:52:31Z

I think it's related to #6041

Yes. It seems to be not resulted by this refactor PR.

wangxiyuan · 2026-01-25T07:26:12Z

you can rebase now.

gcanlin · 2026-01-25T10:14:55Z

@wangxiyuan Ready now.

github-actions · 2026-01-26T01:05:50Z

This pull request has conflicts, please resolve those before we can evaluate the pull request.

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

…rocess` (vllm-project#6191) ### What this PR does / why we need it? Align with upstream vLLM. This PR will help downstream vLLM-Omni reduce the cost for maintaining the _prepare_inputs. Besides, it helps vLLM-Ascend code more readable. In the future, we can follow closer to vLLM. The `preprocess` logic is same as GPUModelRunner. We don't need to maintain it anymore. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? CI. - vLLM version: v0.14.0 - vLLM main: vllm-project/vllm@d682094 --------- Signed-off-by: gcanlin <canlinguosdu@gmail.com>

…rocess` (vllm-project#6191) ### What this PR does / why we need it? Align with upstream vLLM. This PR will help downstream vLLM-Omni reduce the cost for maintaining the _prepare_inputs. Besides, it helps vLLM-Ascend code more readable. In the future, we can follow closer to vLLM. The `preprocess` logic is same as GPUModelRunner. We don't need to maintain it anymore. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? CI. - vLLM version: v0.14.0 - vLLM main: vllm-project/vllm@d682094 --------- Signed-off-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: zrj026 <zhangrunjiang026@gmail.com>

…rocess` (vllm-project#6191) ### What this PR does / why we need it? Align with upstream vLLM. This PR will help downstream vLLM-Omni reduce the cost for maintaining the _prepare_inputs. Besides, it helps vLLM-Ascend code more readable. In the future, we can follow closer to vLLM. The `preprocess` logic is same as GPUModelRunner. We don't need to maintain it anymore. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? CI. - vLLM version: v0.14.0 - vLLM main: vllm-project/vllm@d682094 --------- Signed-off-by: gcanlin <canlinguosdu@gmail.com>

…rocess` (vllm-project#6191) ### What this PR does / why we need it? Align with upstream vLLM. This PR will help downstream vLLM-Omni reduce the cost for maintaining the _prepare_inputs. Besides, it helps vLLM-Ascend code more readable. In the future, we can follow closer to vLLM. The `preprocess` logic is same as GPUModelRunner. We don't need to maintain it anymore. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? CI. - vLLM version: v0.14.0 - vLLM main: vllm-project/vllm@d682094 --------- Signed-off-by: gcanlin <canlinguosdu@gmail.com> Signed-off-by: zrj026 <zhangrunjiang026@gmail.com>

…rocess` (vllm-project#6191) ### What this PR does / why we need it? Align with upstream vLLM. This PR will help downstream vLLM-Omni reduce the cost for maintaining the _prepare_inputs. Besides, it helps vLLM-Ascend code more readable. In the future, we can follow closer to vLLM. The `preprocess` logic is same as GPUModelRunner. We don't need to maintain it anymore. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? CI. - vLLM version: v0.14.0 - vLLM main: vllm-project/vllm@d682094 --------- Signed-off-by: gcanlin <canlinguosdu@gmail.com>

[Refactor] Separate _prepare_inputs to _prepare_inputs and _preprocess

5fd9ed9

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

gcanlin requested a review from MengqingCao as a code owner January 23, 2026 08:12

wangxiyuan added ready read for review ready-for-test start test by label for PR labels Jan 23, 2026

gemini-code-assist Bot reviewed Jan 23, 2026

View reviewed changes

Comment thread vllm_ascend/worker/model_runner_v1.py Outdated

gcanlin added 2 commits January 23, 2026 08:19

remove todo

57c7f5c

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

fix

812a044

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

wangxiyuan added this to the v0.14.0rc1 milestone Jan 24, 2026

wangxiyuan mentioned this pull request Jan 24, 2026

[Release]: Release checklist for v0.14.0rc1 #6149

Closed

38 tasks

github-actions Bot added the merge-conflicts label Jan 24, 2026

github-actions Bot removed the merge-conflicts label Jan 24, 2026

Merge branch 'main' into preprocess

94813ee

gcanlin force-pushed the preprocess branch from 73f512b to 94813ee Compare January 25, 2026 07:27

wangxiyuan approved these changes Jan 25, 2026

View reviewed changes

github-actions Bot added the merge-conflicts label Jan 26, 2026

merge main and fix conflit

1b7d894

Signed-off-by: gcanlin <canlinguosdu@gmail.com>

github-actions Bot removed the merge-conflicts label Jan 26, 2026

wangxiyuan merged commit 6528967 into vllm-project:main Jan 26, 2026
20 checks passed

lidenghui1110 mentioned this pull request Jan 27, 2026

[Bugfix]: fix pp errors when applying flashcomm1 #6282

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Refactor] Separate `_prepare_inputs` to `_prepare_inputs` and `_preprocess`#6191

[Refactor] Separate `_prepare_inputs` to `_prepare_inputs` and `_preprocess`#6191
wangxiyuan merged 5 commits intovllm-project:mainfrom
gcanlin:preprocess

gcanlin commented Jan 23, 2026 •

edited by github-actions Bot

Loading

Uh oh!

github-actions Bot commented Jan 23, 2026

Uh oh!

gemini-code-assist Bot left a comment

Uh oh!

Uh oh!

gcanlin commented Jan 23, 2026

Uh oh!

github-actions Bot commented Jan 24, 2026

Uh oh!

wangxiyuan commented Jan 25, 2026

Uh oh!

wangxiyuan commented Jan 25, 2026

Uh oh!

gcanlin commented Jan 25, 2026

Uh oh!

wangxiyuan commented Jan 25, 2026

Uh oh!

gcanlin commented Jan 25, 2026

Uh oh!

github-actions Bot commented Jan 26, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

gcanlin commented Jan 23, 2026 • edited by github-actions Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What this PR does / why we need it?

Does this PR introduce any user-facing change?

How was this patch tested?

Uh oh!

github-actions Bot commented Jan 23, 2026

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Uh oh!

gcanlin commented Jan 23, 2026

Uh oh!

github-actions Bot commented Jan 24, 2026

Uh oh!

wangxiyuan commented Jan 25, 2026

Uh oh!

wangxiyuan commented Jan 25, 2026

Uh oh!

gcanlin commented Jan 25, 2026

Uh oh!

wangxiyuan commented Jan 25, 2026

Uh oh!

gcanlin commented Jan 25, 2026

Uh oh!

github-actions Bot commented Jan 26, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

gcanlin commented Jan 23, 2026 •

edited by github-actions Bot

Loading