[Test] Add BuildKite test-full script for full CI.#867
[Test] Add BuildKite test-full script for full CI.#867yenuo26 wants to merge 7 commits intovllm-project:mainfrom
Conversation
Signed-off-by: wangyu31577 <wangyu31577@hundsun.com>
| engine_output_type: latent # Output hidden states for talker | ||
| distributed_executor_backend: "mp" | ||
| max_num_batched_tokens: 32768 | ||
| max_num_batched_tokens: 30000 |
There was a problem hiding this comment.
why change this number?
There was a problem hiding this comment.
The modifications have been deleted.
Signed-off-by: wangyu31577 <wangyu31577@hundsun.com>
|
@hsliuustc0106 @david6666666 Please add a ready label, and merge this PR ASAP. We can enable nightly daily-CI after this PR. |
|
@hsliuustc0106 This PR is the pre-work for the nightly CI build. PTAL. |
congw729
left a comment
There was a problem hiding this comment.
LGTM. We can take the Qwen3omni expansion test cases to build the nightly CI at first, and add more tests afterwards.
There was a problem hiding this comment.
Pull request overview
This PR adds a new BuildKite CI configuration file for running comprehensive tests. The configuration is designed to test Qwen3-Omni model expansion functionality on H100 GPUs.
Changes:
- Added
.buildkite/test-full.yamlwith a two-step pipeline: Docker image build and H100 GPU testing - The configuration runs the
test_qwen3_omni_expansion.pytest file which contains multiple test cases for text-to-text, audio-to-text, audio-to-text-audio, image-to-text, and image-to-text-audio scenarios
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Signed-off-by: wangyu31577 <wangyu31577@hundsun.com>
|
@hsliuustc0106 PTAL |
|
@hsliuustc0106 PTAL |
PLEASE FILL IN THE PR DESCRIPTION HERE ENSURING ALL CHECKLIST ITEMS (AT THE BOTTOM) HAVE BEEN CONSIDERED.
Purpose
Add BuildKite test-full script for full CI.
Test Plan
Submit a debug PR and run the corresponding script in the pipeline.
Test Result
Debug PR: #849

Essential Elements of an Effective PR Description Checklist
supported_models.mdandexamplesfor a new model.BEFORE SUBMITTING, PLEASE READ https://github.com/vllm-project/vllm-omni/blob/main/CONTRIBUTING.md (anything written below this line will be removed by GitHub Actions)