[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
+4 -2
View File
@@ -28,8 +28,10 @@ MODEL_QUANT = [
@pytest.mark.parametrize("model_id, use_marlin_kernel", MODEL_QUANT)
def test_gptq_with_dynamic(vllm_runner, model_id: str,
use_marlin_kernel: bool):
def test_gptq_with_dynamic(vllm_runner, model_id: str, use_marlin_kernel: bool,
monkeypatch):
# vllm_runner.apply_model() relies on V0 internals.
monkeypatch.setenv("VLLM_USE_V1", "0")
vllm_model = vllm_runner(model_id, dtype=torch.float16, max_model_len=2048)