Skip to content

Comments

[Refactor] Simplify dummy data generation#35025

Open
DarkLight1337 wants to merge 8 commits intovllm-project:mainfrom
DarkLight1337:simplify-dummy-data
Open

[Refactor] Simplify dummy data generation#35025
DarkLight1337 wants to merge 8 commits intovllm-project:mainfrom
DarkLight1337:simplify-dummy-data

Conversation

@DarkLight1337
Copy link
Member

@DarkLight1337 DarkLight1337 commented Feb 21, 2026

Purpose

Test Plan

Using the same test as #34330

Test Result

(Worker_TP2 pid=1233399) INFO 02-21 15:00:08 [gpu_model_runner.py:5150] Encoder cache will be initialized with a budget of 114688 tokens, and profiled with 1 video items of the maximum feature size.
(Worker_TP3 pid=1233400) INFO 02-21 15:00:08 [gpu_model_runner.py:5150] Encoder cache will be initialized with a budget of 114688 tokens, and profiled with 1 video items of the maximum feature size.
(Worker_TP1 pid=1233398) INFO 02-21 15:00:08 [gpu_model_runner.py:5150] Encoder cache will be initialized with a budget of 114688 tokens, and profiled with 1 video items of the maximum feature size.
(Worker_TP0 pid=1233397) INFO 02-21 15:00:08 [gpu_model_runner.py:5150] Encoder cache will be initialized with a budget of 114688 tokens, and profiled with 1 video items of the maximum feature size.

Essential Elements of an Effective PR Description Checklist
  • The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
  • The test plan, such as providing test command.
  • The test results, such as pasting the results comparison before and after, or e2e results
  • (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.
  • (Optional) Release notes update. If your change is user facing, please update the release notes draft in the Google Doc.

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
@dosubot
Copy link

dosubot bot commented Feb 21, 2026

Related Documentation

Checked 0 published document(s) in 1 knowledge base(s). No updates required.

How did I do? Any feedback?  Join Discord

@mergify
Copy link

mergify bot commented Feb 21, 2026

Documentation preview: https://vllm--35025.org.readthedocs.build/en/35025/

@mergify mergify bot added documentation Improvements or additions to documentation deepseek Related to DeepSeek models llama Related to Llama models multi-modality Related to multi-modality (#4194) qwen Related to Qwen models speculative-decoding labels Feb 21, 2026
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request refactors the dummy data generation for multimodal models by centralizing the merging of keyword arguments from the multimodal configuration. This simplifies the get_dummy_mm_data and get_hf_processor methods across numerous model implementations. While the refactor is comprehensive, there are a few inconsistencies in the updated method signatures and missing keyword argument support in some custom processor adapters that could lead to runtime errors.

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

deepseek Related to DeepSeek models documentation Improvements or additions to documentation llama Related to Llama models multi-modality Related to multi-modality (#4194) qwen Related to Qwen models ready ONLY add when PR is ready to merge/full CI is needed speculative-decoding

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants