EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[CI] Fix ngram & suffix test oom (#4755)

Browse Source 
### What this PR does / why we need it?
Avoid oom during CI by using `with VllmRunner` instead of `LLM()`, and
enable `test_ngram_correctness`

### How was this patch tested?
CI passed.

- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

---------

Signed-off-by: fluctlux <38945811+fluctlux@users.noreply.github.com>
Co-authored-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
fluctlux
2025-12-08 09:26:29 +08:00
committed by GitHub
parent 916a9a1913
commit 9fbcfa36af
2 changed files with 13 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
.github/workflows/_e2e_test.yaml vendored
View File

@@ -108,8 +108,7 @@ jobs:
# ------------------------------------ v1 spec decode test ------------------------------------ #
pytest -sv tests/e2e/singlecard/spec_decode_v1/test_v1_mtp_correctness.py
pytest -sv tests/e2e/singlecard/spec_decode_v1/test_v1_mtp_torchair_correctness.py
# Fix me: test_eagle_correctness OOM error
#pytest -sv tests/e2e/singlecard/spec_decode_v1/test_v1_spec_decode.py
pytest -sv tests/e2e/singlecard/spec_decode_v1/test_v1_spec_decode.py
e2e-2-cards:
name: multicard-2