A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 13, 17:00 UTC
[CI/Build] Loosen STT LoRA Translate Check (Flaky Test) (#28247)
Signed-off-by: Alex-Brooks <Alex.Brooks@ibm.com> Signed-off-by: Alex Brooks <alex.brooks@ibm.com> Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
Alex Brooks committed
Nov 6, 2025 at 19:51 UTC
e70fbc599b8e0306b093b51aa11d13957d7d00a8
1 file changed
+1
-1
tests/entrypoints/openai/test_translation_validation.py
+1
-1
@@ -79,7 +79,7 @@ async def test_basic_audio_with_lora(mary_had_lamb):
79
temperature=0.0,
80
)
81
out = json.loads(translation)["text"].strip().lower()
82
- assert "mary tenía un pequeño cordero" in out
82
+ assert "pequeño" in out.split(" ")
83
84
85
# NOTE: (NickLucche) the large-v3-turbo model was not trained on translation!