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 16, 03:15 UTC
[CI] Turn on basic correctness tests for V1 (#10864)
Tyler Michael Smith committed
Jan 8, 2025 at 21:20 UTC
615e4a54017136649db275b68932af80168781f8
1 file changed
-1
tests/basic_correctness/test_basic_correctness.py
-1
@@ -44,7 +44,6 @@ def test_vllm_gc_ed():
44
assert weak_llm() is None
45
46
47
-@pytest.mark.skip_v1
47
@pytest.mark.parametrize("model", MODELS)
48
@pytest.mark.parametrize("backend", ["FLASH_ATTN", "XFORMERS", "FLASHINFER"])
49
@pytest.mark.parametrize("dtype", ["half"])