[V1] V1 Enablement Oracle (#13726)

Signed-off-by: rshaw@neuralmagic.com <rshaw@neuralmagic.com>
Co-authored-by: rshaw@neuralmagic.com <rshaw@neuralmagic.com>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>
Co-authored-by: Michael Goin <michael@neuralmagic.com>
This commit is contained in:
Robert Shaw
2025-03-14 22:02:20 -07:00
committed by GitHub
co-authored by rshaw@neuralmagic.com <rshaw@neuralmagic.com> Nicolò Lucchesi Tyler Michael Smith Michael Goin
parent 8c0d15d5c5
commit d4d93db2c5
96 changed files with 1537 additions and 512 deletions
+3 -1
View File
@@ -10,7 +10,9 @@ from vllm.model_executor.layers.quantization.quark.quark import ( # noqa: E501
QuarkLinearMethod, QuarkW8A8Fp8)
def test_quark_fp8(vllm_runner):
def test_quark_fp8(vllm_runner, monkeypatch):
# vllm_runner.apply_model() relies on V0 internals.
monkeypatch.setenv("VLLM_USE_V1", "0")
model_path = "amd/Llama-3.1-8B-Instruct-FP8-KV-Quark-test"
with vllm_runner(model_path) as llm: