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 15, 07:15 UTC
[Quantization] Skip Fp4 Test for `compressed-tensors` (#19217)
Dipika Sikka committed
Jun 5, 2025 at 14:21 UTC
aa49f148322a39727be110da51a6782d43a2f5d8
1 file changed
+1
tests/quantization/test_compressed_tensors.py
+1
@@ -651,6 +651,7 @@ def test_compressed_tensors_2of4_sparse_compressed(vllm_runner, args_2of4):
651
assert output
652
653
654
+@pytest.mark.skip(reason="Skip until the model config is updated")
655
def test_compressed_tensors_nvfp4a16(vllm_runner):
656
# run weight only example
657
model = "nm-testing/TinyLlama-1.1B-Chat-v1.0-FP4"