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 14, 05:30 UTC
[CI/Build] Disable flaky test_structured_output tests (#24404)
Signed-off-by: 22quinn <33176974+22quinn@users.noreply.github.com>
22quinn committed
Sep 7, 2025 at 19:51 UTC
3a3e91bdfe86adcb8fceb1cb8c5f883878fc65b4
1 file changed
+4
-4
tests/v1/entrypoints/llm/test_struct_output_generate.py
+4
-4
@@ -46,12 +46,12 @@ PARAMS_MODELS_BACKENDS_TOKENIZER_MODE = [
46
("mistralai/Ministral-8B-Instruct-2410", "xgrammar", "mistral", None),
47
("Qwen/Qwen2.5-1.5B-Instruct", "xgrammar", "auto", None),
48
("Qwen/Qwen2.5-1.5B-Instruct", "lm-format-enforcer", "auto", None),
49
- ("mistralai/Ministral-8B-Instruct-2410", "outlines", "auto", None),
50
- ("mistralai/Ministral-8B-Instruct-2410", "outlines", "mistral", None),
49
+ #FIXME: This tests are flaky on CI thus disabled. Tracking in Issue #24402
50
+ # ("mistralai/Ministral-8B-Instruct-2410", "outlines", "auto", None),
51
+ # ("mistralai/Ministral-8B-Instruct-2410", "outlines", "mistral", None),
52
+ #("Qwen/Qwen2.5-1.5B-Instruct", "guidance", "auto"),
53
("mistralai/Ministral-8B-Instruct-2410", "outlines", "auto",
54
NGRAM_SPEC_CONFIG),
53
- #FIXME: This test is flaky on CI thus disabled
54
- #("Qwen/Qwen2.5-1.5B-Instruct", "guidance", "auto"),
55
("mistralai/Ministral-8B-Instruct-2410", "guidance", "auto",
56
NGRAM_SPEC_CONFIG),
57
("Qwen/Qwen2.5-1.5B-Instruct", "xgrammar", "auto", NGRAM_SPEC_CONFIG),